AN ABSTRACT OF THE DISSERTATION OF

advertisement
AN ABSTRACT OF THE DISSERTATION OF
Kimberly J. Hageman for the degree of Doctor of Philosophy in Chemistry presented on
14. 2003.
Title: Measuring In Situ Reductive Dechlorination Rates in Trichloroethene-
Contaminated Groundwater.
Abstract approved:
Redacted for Privacy
Jennifer A. Field
Trichloroethene (TCE) is the most frequently detected organic contaminant in groundwater, is
classified as a probable human carcinogen, and exhibits toxicological effects on the human
endocrine, immune, developmental, and reproductive systems.
While significant research
efforts have been devoted to the development of strategies for remediating TCE-contaminated
groundwater, their advancement is currently hindered by limitations in current methodologies
for measuring in situ reductive dechlorination rates, especially for sorbing solutes.
This
dissertation describes the development, evaluation, and demonstration of a method for
measuring in situ reductive dechlorination rates that utilizes single-well, "push-pull" test
technology.
Initial field tests indicated that trichlorofluoroethene (TCFE) could be used as a
surrogate for TCE in push-pull tests since (a) TCE and TCFE were transported similarly and
(b) TCFE underwent reductive dechlorination by a pathway analogous to that of TCE while
retaining the fluorine label. Because TCFE and TCE experienced sorption at the selected field
site, a novel data analysis technique called "forced mass balance" (FMB) was developed to
obtain in situ transformation rates of sorbing solutes from push-pull test data. The FMB
technique was evaluated by quantifying errors in rates derived by applying FMB to push-pull
test data generated by a numerical model.
Results from simulated tests indicated that an
example in situ rate for the reductive dechlorination of TCFE, which was obtained by applying
FMB to field data, was underestimated relative to the true in situ rate by 10%. The utility of
the rate-determination method presented in this dissertation was demonstrated by using it to
evaluate the effectiveness of a chemical amendment, namely fumarate, at enhancing in situ
reductive dechlorination rates in TCE-contaminated groundwater.
Reductive dechlorination
rates increased following three consecutive additions of fumarate in all five of the tested wells.
The development of the rate-determination method described in this dissertation advances the
state of bioremediation technology because methods for measuring in situ transformation rates
are needed to both assess the potential for natural attenuation and to quantify the effects of
bioremediation techniques in the field.
© Copyright by Kimberly J. Hageman
April 14, 2003
All Rights Reserved
MEASURING IN SITU REDUCTIVE DECHLORINATION RATES IN
TRICHLOROETI{ENE-CONTAMINATED GROUNDWATER
by
Kimberly J. Hageman
A DISSERTATION
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
Presented April 14, 2003
Commencement June 2003
Doctor of Philosophy dissertation of Kimberly J. Hageman presented on April 14, 2003.
APPROVED:
Redacted for Privacy
Major Pro'resentin Chemistry
Redacted for Privacy
Head of Drtment of Chemistry
Redacted for Privacy
Dean of the GratklatèSchool
I understand that my dissertation will become part of the permanent collection of Oregon State
University libraries. My signature below authorizes release of my dissertation to any reader
upon request.
Redacted for Privacy
ieman Author
ACKNOWLEDGEMENTS
The author expresses sincere appreciation to Rob, Lynn, Vickie, Kara, and Katie for their
support and understanding over the years. In addition, the author is grateful to her major
professors, Drs. Jennifer Field and Jonathan Istok, for providing guidance, encouragement,
and countless charades clues. Other individuals who provided valued assistance include Ralph
Reed, Brian Davis, Jesse Jones, Kevin Tam, Brian Wood, Roy Haggerty, Mark Dolan,
Angelito Tirona, Kirk O'Reilley, Mark White, Melora Park, Mark Humphrey, and Andrew
Lundmark.
The Western Regions Hazardous Substance Research Center, the National
Institute of Environmental Health Sciences' Superfund Basic Research Program, and the
ChevronTexaco Energy Research and Technology Company provided financial support.
CONTRIBUTION OF AUTHORS
Drs. Jennifer A. Field and Jonathan D. Istok provided guidance in all aspects of this
dissertation. Dr. Lewis Semprini provided guidance in designing and interpreting results from
field experiments and edited chapters 2 and 4. Dr. Timothy E. Buscheck facilitated the field
campaigns and edited chapter 2. Dr. Martin H. Schroth provided guidance in designing and
interpreting computer modeling experiments and edited chapter 3.
TABLE OF CONTENTS
CHAPTER L INTRODUCTION .............................................................................................
CHAPTER 2. IN SITU ANAEROBIC TRANSFORMATION OF
TRTCHLOROFLUOROETHENE IN TRICHLOROETHENECONTAMINATED GROUNDWATER........................................................... 9
ABSTRACT ................................................................................................................ 10
INTRODUCTION ....................................................................................................... 11
EXPERIMENTAL SECTION .................................................................................... 13
Chemicals ........................................................................................................ 13
SiteDescription ............................................................................................... 13
Push-Pull Tests ................................................................................................ 14
AnalyticalMethods ......................................................................................... 17
RESULTS & DISCUSSION ....................................................................................... 18
TransportTests ................................................................................................ 18
Transformation Tests ....................................................................................... 20
ACKNOWLEDGEMENTS ........................................................................................ 29
LITERATURE CITED ............................................................................................... 29
CHAPTER 3. "FORCED MASS BALANCE" TECHNIQUE FOR ESTIMATING IN
SITU TRANSFORMATION RATES OF SORB1NG SOLUTES IN
GROUNDWATER .......................................................................................... 31
ABSTRACT ................................................................................................................ 32
INTRODUCTION ....................................................................................................... 33
METHODS................................................................................................................. 36
Numerical Simulations .................................................................................... 36
Application of FMB to a Field Push-Pull Test ................................................ 37
RESULTS AND DISCUSSION ................................................................................. 38
TestSetl .......................................................................................................... 38
TestSet II ........................................................................................................ 42
TestSet III ....................................................................................................... 43
FIELD APPLICATION .............................................................................................. 45
ACKNOWLEDGEMENTS ........................................................................................ 46
LITERATURE CITED ............................................................................................... 46
SUPPORTING INFORMATION: Example input file for use with STOMP ............ 47
TABLE OF CONTENTS (Continued)
CHAPTER 4. QUANTIFYING THE EFFECTS OF CHEMICAL AMENDMENTS
ON IN SITU REDUCTIVE DECHLORINATION RATES .......................... 51
ABSTRACT ................................................................................................................ 52
INTRODUCTION ....................................................................................................... 53
EXPERIMENTAL SECTION
....................................................................................
56
Chemicals ........................................................................................................ 56
SiteDescription ............................................................................................... 56
Push-Pull Tests ................................................................................................ 56
Analytical Methods ......................................................................................... 57
DataAnalysis .................................................................................................. 59
RESULTS AND DISCUSSION ................................................................................. 60
Reductive Dechlorination Rates and Product Distribution Ratios .................. 60
Correlations between Fumarate and TCFE Transformation Behavior
66
............
CONCLUSIONS ......................................................................................................... 68
ACKNOWLEDGEMENTS ........................................................................................ 69
LITERATURE CITED
...............................................................................................
69
CHAPTER 5: SUMMARY ...................................................................................................... 71
LIST OF FIGURES
Figure
2.1. Reductive dechlorination pathways for (a) TCE (5) and (b) TCFE (12). The
predominant pathways are indicated by underlines and heavy arrows ................
11
2.2. Experimental set-up for test solution injection in (a) the A-zone well and (b) C-
zone wells (not drawn to scale) .........................................................................
16
2.3. Breakthrough curves indicating (a) non-conservative transport of TCFE and TCE
in A-zone well RI-bA (test 1) and (b) conservative transport of TCFE
and DCFE isomers in C-zone well GW-21C (test 2) ......................................... 19
2.4. Test 4 data indicating (a) changes in concentration for formate and relative
concentration for bromide and (b) reductive dechlorination of injected
TCFE to cis-DCFE, trans-DCFE, and CFE in A-zone well RI-bA.
Measured concentrations are adjusted for dilution ............................................. 22
2.5. Reductive dechlorination of (a) injected TCFE to cis-DCFE, trans-DCFE, and
CFE and (b) TCE to cis-DCE, trans-DCE, and 1,1-DCE in A-zone well
RI-bA with added formate as an exogenous electron donor (test 4).
Measured concentrations are adjusted for dilution ............................................. 24
2.6. Reductive dechlorination of injected TCFE to cis-DCFE, trans-DCFE and CFE
in A-zone well RI-i OA without added formate as an exogenous electron
donor (test 5). Meausured concentrations are adjusted for dilution ................... 25
2.7. Reductive dechlorination of injected TCFE to cis- and trans-DCFE occurring
after 55 days in C-zone well GW- 15 C with added formate as an
exogenous electron donor (test 6). Measured concentrations are adjusted
fordilution .......................................................................................................25
2.8. Absence of reductive dechlorination of injected TCFE in C-zone well GW-21C
without added formate as an exogenous electron donor in groundwater
that had not previously been exposed to TCE or other co-contaminants
(test 7). Measured concentrations are adjusted for dilution ............................... 27
3.1. Test Set I, Test 1,
R(A) = R(B) = 5. Total (aqueous plus sorbed) concentrations of
the reactant, A, the product, B, and the tracer, T, in the simulated aquifer
at (a) the end of the test solution injection (125 mm) and (b) 90 days.
The vertical line (at 0 days) represents the single well. (c) Total
concentrations and (d) FMB- and ideal transport-adjusted concentrations
at the simulated well ......................................................................................... 39
3.2. Test Set I, Test 2, R(A) = 5, R(B) = 1.25. (a) Total concentrations of A, B, and T
in the simulated aquifer at 90 days. (b) FMB- and ideal transportadjusted concentrations at the simulated well.................................................... 40
LIST OF FIGURES (Continued)
Figure
3.3.
3.4.
3.5.
3.6.
4.1.
4.2.
4.3.
Pgc
Test Set I, Test 3, R(A) = 5, R(B) = 20. (a) Total concentrations of A, B, and T in
the simulated aquifer at 90 days. (b) FMB- and ideal transport-adjusted
concentrations at the simulated well ................................................................. 41
Test Set II. Error quotients as a function of (a) test duration, (b) groundwater
velocity, (c) input k, (d) dispersivity ................................................................. 43
Test Set III. (a) Error quotients as they relate to retardation factor when R(A)>
R(B). (b) Total concentrations of A, B, and T in the simulated aquifer at
90 days when R(A) 50 and R(B) = 1. (c) Error quotients as they relate
to retardation factor when R(A) <R(B). (d) Total concentrations of A, B,
and T in the simulated aquifer at 90 days when R(A) = 5 and R(B) = 250 .......... 44
(a) FMB-adjusted concentrations of TCFE and its products, indicating that
reductive dechlorination of TCFE occurred during this field push-pull
test conducted in TCE-contaminated groundwater and (b) the plot from
which the derived first-order rate constant was determined ............................... 45
Analogous reductive dechlorination pathways for (a) TCE and (b) its fluorinated
surrogate, TCFE. Predominant isomers and pathways are indicated by
underlines and heavy arrows ............................................................................. 53
Reduction of fumarate to succinate ............................................................................... 54
Test 1 in Well 1 OA. (a) Aqueous measured concentrations illustrating
concentration changes due to transformation and transport processes, (b)
forced mass balance (FMB)-adjusted concentrations (see data analysis
section) illustrating concentration changes due to transformation only
and (c) the plot used to determine the first-order rate constant (k) ..................... 61
4.4. Stacked area plots of FMB-concentrations depicting reductive dechlorination of
TCFE following the first (test 1) and second (test 5) injections of TCFE
inA-zone wells ................................................................................................ 63
4.5.
4.6.
Stacked area plots of F MB-concentrations depicting reductive dechlorination of
TCFE following the first (test 1) and second (test 5) injections of TCFE
inC-zone wells ................................................................................................ 64
Tracer-normalized (TN) concentrations (see data analysis section) of fumarate
and its reduction product, succinate .................................................................. 66
LIST OF TABLES
Table
2.1. Contaminants and Biogeochemical Indicators in Selected Wells ................................... 14
2.2. Push-Pull Test Descriptions .......................................................................................... 15
2.3. Reductive Dechlorination Rates for TCFE, TCE, and Their Transformation
Products in Push-Pull Transformation Tests ..................................................... 26
4.1. Test Solution Compositions .......................................................................................... 58
4.2. TCFE Transformation Rates ......................................................................................... 62
MEASURING IN SITU REDUCTIVE DECHLORINATION RATES IN
TRICHLOROETHENE-CONTAMINATED GROUNDWATER
CHAPTER 1. INTRODUCTION
Trichioroethene (TCE) is a synthetic chlorinated hydrocarbon known for its solvent
properties and low fire and explosion potential. TCE has been used extensively since the
1920's for dry cleaning, degreasing fabricated metal parts, and as a solvent for fats, waxes,
resins, oils, rubber, paints, and varnishes since the 1920's (1). Due to the common occurrence
of leaks, spills, and poor disposal practices at industrial sites, significant volumes of TCE have
been released to the environment. Released TCE tends to volatilize to the atmosphere since its
Henry's Law constant and boiling point are 0.020 atm-m3/mol at 20 °C and 88
C,
respectively. However, because TCE is water soluble (1.1 g/L at 20 C) and it does not sorb
strongly to soils, much of the TCE that does not volatilize migrates into groundwater (2).
Because the density of TCE (1.46 g/mL at 20 C) is greater than that of water, TCE tends to
migrate deep beneath the water table, contaminating all groundwater that it contacts. TCE is
the most frequently detected organic contaminant in groundwater (3) and the most frequently
detected chemical at Superfund sites (4).
While TCE lowers the risk for fires and explosions at industrial sites and therefore
saves human lives, another type of risk to human health is created by the contamination of
potential drinking water sources with TCE (5). The primary public health concern associated
with TCE is cancer induction through chronic exposure to low levels of TCE in drinking
water (1).
While TCE itself is not suspected to be carcinogenic, the products of TCE
metabolism in mammals are blamed for its mutagenic effects (5). Besides carcinogenicity,
TCE exhibits toxicological effects on the endocrine, immune, developmental, and
reproductive systems (1).
The maximum contaminant level (MCL), or the lowest
concentration to which a contaminant can be removed from drinking water using current
technology, for TCE is 5 ppb (6). However, the California EPA recommends that the state
MCL be lowered to 0.8 ppb based on results from recent toxicological studies (1).
Widespread concern about the toxicological effects of TCE has driven research
activity to focus on the development of strategies for remediating TCE-contaminated
2
groundwater. As recently as 1980, TCE was thought to be non-degradable in groundwater (2,
7) and pump-and-treat was considered the only viable remediation strategy (8). However,
pump-and-treat, which involves pumping contaminated groundwater to the surface for
treatment, has been generally ineffectual (9-11). Moreover, surface treatment processes, such
as air stripping or carbon adsorption, simply transfer TCE to another medium instead of
destroying it (9). In the mid-1980's, investigators began to report the occurrence of TCE
transformation products in groundwater (2). These findings inspired a new research effort
directed towards understanding possible mechanisms of TCE transformation in groundwater.
Since then, several reviews have been published that describe potential TCE transformation
mechanisms by both abiotic and biological processes (2, 5, 8, 12, 13).
The average oxidation state of the carbon atoms in TCE is positive due to the presence
of the three electronegative chlorine atoms in the molecule. As a result, these carbon atoms do
not have electrons available to donate, which makes it difficult for microorganisms to obtain
energy from TCE via oxidation. For this reason, TCE is recalcitrant in aerobic aquifers
relative to other common contaminants, from which microorganisms can gain energy by
mediating the transfer of their available electrons to oxygen. Although no microorganism is
known to exist that can use TCE as its sole electron source (14, 15), TCE can be oxidized by
aerobic organisms through cometabolism (2, 5, 14). The aerobic cometabolism of TCE relies
on the fortuitous oxidation of TCE by enzymes produced by microorganisms for other
purposes. Concerted research efforts are underway to better understand the mechanisms that
drive the transformation of TCE by aerobic cometabolism and to promote the transformation
of TCE
by aerobic cometabolism in remediation projects (16-18).
In anaerobic aquifer environments, the predominant pathway for TCE transformation
is reductive dechlorination (8, 13, 15, 19). In this reaction, microorganisms gain energy by
mediating the transfer of electrons from electron donors to TCE.
In the reductive
dechlorination pathway, hydrogen atoms replace chlorine atoms in the TCE molecule thereby
driving the sequential reduction of TCE to the dichioroethene (DCE) isomers, chioroethene
(CE), and ethene.
Among the pure culture strains capable of facilitating the reductive
dechlorination of TCE are Dehalospirillum multivorans (20), Dehalococcoides ethenogenes
Strain 195 (21), Desuijitobacterium strain PCE-S (22), and Dehalobacter restrictus (23).
However, none of these pure cultures are capable of reducing TCE to the completely
dechlorinated product, ethene, which is the desired end product of reductive dechlorination
3
due to its lack of toxicity. It is especially concerning when reductive dechlorination results in
the accumulation of CE since CE is most toxic of the chemicals in the TCE reductive
dechlorination pathway.
However, microbial communities composed of syntrophically-
associated microorganisms are capable of performing complete reductive dechlorination (2426).
Based on the wealth of information that has been obtained about the reductive
dechlorination pathway, in situ bioremediation via reductive dechlorination has become a
predominant strategy for remediating TCE-contaminated groundwater (27, 28). The two
primary management approaches associated with in situ bioremediation are "monitored
intrinsic bioremediation" and "engineered bioremediation." Intrinsic bioremediation relies on
indigenous microorganisms to reduce contaminant mass without human intervention. The
natural attenuation of TCE by reductive dechlorination occurs when dechlorinating microbial
communities co-exist with co-contaminants or naturally-occurring organic compounds that
can act as electron donors (29). The use of intrinsic bioremediation as a management strategy
requires proof that contaminant loss due to transformation is occurring (30). Additionally,
intrinsic bioremediation protocols often require that computer modeling be used to predict the
future migration and attenuation of contaminant plumes (28). To obtain accurate predictions,
site-specific rates for in situ transformation must be obtained so that they can be used as input
values in solute fate and transport models.
Where natural attenuation does not result in the complete conversion of TCE to ethene
or where transformation rates are too slow to meet risk management goals, engineered
bioremediation approaches are needed. A common engineered approach to enhancing in situ
reductive dechlorination rates is to stimulate the growth of indigenous dechlorinating
microorganisms by adding chemical amendments to groundwater.
A wide variety of
chemicals and chemical mixtures have been evaluated for their suitability as amendments for
enhancing reductive dechlorination. Lee et al. reviewed results from laboratory tests that were
designed to assess the effectiveness of potential amendments such as complex organic
mixtures (molasses, wastewater, cheese whey permeate, corn steep liquor, manure tea),
metabolic intermediates (benzoate, lactate, propionate, acetate, butyrate), alcohols (methanol,
ethanol), molecular hydrogen, sulfate, nitrate, vitamins, and micronutrients (29, 31). While
many of these amendments were effective, disadvantages were associated with each and none
were universally effective. The effectiveness of a chemical amendment is generally evaluated
4
by comparing reductive dechlorination rates measured with and without the chemical
amendment in laboratory experiments with pure or mixed cultures of microorganisms.
However, due to potential discrepancies between laboratory and field results (28, 32, 33), it is
also necessary to evaluate the effects of chemical amendments on in situ reductive
dechlorination measured during field tests.
While protocols for remediating TCE-contaminated groundwater by intrinsic or
engineered bioremediation require methods for measuring in situ transformation rates, in situ
transformation rates are difficult to measure. In situ transformation rates are difficult to
measure because solute concentrations in groundwater are affected by both transformation and
transport processes (27 28, 34). For example, solute concentrations measured at a single well
change with time due to advection, which is the process by which solutes are transported with
bulk groundwater flow. Dispersion is the process by which solute concentrations decrease due
to the broadening of solute plumes as they flow through the porous aquifer medium. There are
three basic processes that cause dispersion (35). First, when a fluid travels though pores, the
velocity of the fluid is faster in the center of the pore than along the edges. Second, some fluid
will follow longer flow paths than other fluid. Third, the velocity of fluid traveling through
larger pores is greater than that flowing through smaller pores. Sorption is the process by
which solutes partition to aquifer solids, thereby decreasing their aqueous concentrations and
slowing their rate of transport in groundwater.
A number of in situ rate-determination methods have been described that involve the
measurement of reactant and product concentrations in multiple wells.
For example,
McAllister and Chiang described the "mass balance approach" (36). In this approach, the total
mass of a reactive solute (reactant) in an aquifer is determined by collecting groundwater
samples from an extensive monitoring well network that encompasses the complete vertical
and horizontal extent of the solute plume. In situ transformation rates are calculated from
changes in the total mass of the solute over time. In the "Technical Protocol for Evaluating
Natural Attenuation of Chlorinated Solvents in Groundwater" published by the United States
EPA in 1998, a "data set normalization" approach for determining in situ transformation rates
was described (28).
This approach relies on the presence of a nonreactive tracer that is
associated with the contaminant plume. Example tracers include trimethylbenzene, which is a
nonreactive component of fuel hydrocarbons; the chloride ion produced during reductive
dechlorination; and the carbon nucleus of chlorinated ethenes.
Reactant and tracer
i!1
concentrations are measured in multiple wells arranged along a transect. The effects of
transport processes are accounted for by normalizing measured reactant concentrations to their
corresponding measured tracer concentrations. Finally, Buscheck and Alcantar presented a
rate-determination method that utilizes a novel analytical solution to the advection-dispersion
equation (37). Reactant concentrations are measured in multiple wells arranged in a transect
and then the analytical solution is used to determine the transformation rate that would be
necessary to produce a steady-state plume of the configuration found at the field site.
However, McNab and Dooher cautioned against the use of this method because it produces
spurious results if the solute plume has not truly reached a steady-state condition (38).
Temporal differences in reactant and product concentrations measured at a single well
can also be used to determine in situ transformation rates. Methods that use single wells tests
are advantageous over those that require multiple wells for a number of reasons. For example,
single-well tests are cost-effective relative to multi-well tests because fewer groundwater
wells, which are expensive to construct, are needed. Additionally, single-well tests do not
require the fortuitous association of a nonreactive tracer with the solute plume or that the
solute plume has reached steady-state. Single-well tests take less time to conduct than multiwell tests since injected solutes do not have to be transported between wells. Because singlewell tests are time-efficient, they can be repeated in a single well to assess reproducibility or to
compare rates obtained with different chemical amendments. At sites where a number of
monitoring wells exist, single-well tests can be conducted simultaneously in different wells to
assess spatial variability.
Washington et al. described a method that utilizes reactant
concentrations measured at a single well and site-specific hydrologic properties with "full
inverse" modeling to differentiate concentration changes due to transformation from those due
to transport processes
(33).
This dissertation describes the development and demonstration of
an alternative approach that uses data collected during single-well "push-pull" tests to
determine in situ transformation rates. Push-pull tests are conducted by injecting ("pushing")
an aqueous test solution containing a conservative tracer and one or more reactants into an
aquifer via a groundwater well
(39).
Samples of the test solution/groundwater mixture are
then extracted ("pulled") from the same well over time and analyzed for tracer, reactant, and
product concentrations.
In chapter 2 of this dissertation, the transport and transformation behavior of
trichlorofluoroethene (TCFE) in a TCE-contaminated aquifer is described.
The TCE-
6
contaminated aquifer was located at the site of a former pesticide-manufacturing site in the
San Francisco Bay area. Because field experiments described in chapter 2 indicated that the
transport and transformation behavior of TCFE was similar to that of TCE, TCFE was used as
a surrogate for TCE in proceeding push-pull tests. TCE itself was not used in push-pull tests
because mixing of the injected test solution with native groundwater would have rendered it
impossible to distinguish injected and background TCE. In chapter 3, a novel data analysis
technique for determining in situ transformation rates of sorbing solutes from push-pull test
data is described. While this data analysis technique can be used with a variety of sorbing
solutes, its development was especially critical to this dissertation because TCFE and TCE
sorb to aquifer materials at the selected field site. In chapter
4,
the utility of the rate-
determination method developed in chapters 2 and 3 is demonstrated by using it to quantify
the effects of a chemical amendment, namely fumarate (trans-i ,2-ethenedicarboxylate), on in
situ transformation rates.
LITERATURE CITED
(1)
California EPA Office of Environmental Health Hazard Assessment "Public Health
Goals for Trichloroethylene in Drinking Water," 1999,
www.oehha.org/water/phglpdf/tce_f.pdf.
(2)
McCarty, P. L.; Semprini, L. In Handbook of Bioremediation; RD Norris, Hinchee,
RE, Brown, R, McCarty, PL, Semprini, L, Wilson, DH, Kampbell, M, Reinhard, EG,
Bouwer, R, Borden, C, Vogel, TM, Thomas, JM, Ward, CH, Eds.; Lewis Publishers:
Boca Raton, 1994; pp 87-116.
(3)
Domenico, P. A.; Schwartz, F. W. Physical and Chemical Hydrogeology; John Wiley
and Sons: New York,
1990.
(4)
Abelson, P. H. Science 1990,
(5)
Ensley, B. D. Ann. Rev. Microbiol. 1991,
(6)
US EPA Office of Groundwater and Drinking Water "Consumer Factsheet on
Trichloroethylene," 2003, www.epa.gov/OGWDW.
(7)
Vogel, T. M. In Handbook of Bioremediation; RD Norris, Hinchee, RE, Brown, R,
McCarty, PL, Semprini, L, Wilson, DH, Kampbell, M, Reinhard, EG, Bouwer, R,
Borden, C, Vogel, TM, Thomas, JM, Ward, CH, Eds.; Lewis Publishers: Boca Raton,
1994; pp
(8)
247, i4i.
45,
283-299.
20 1-225.
Bradley, P. M. Hydrogeol. 1
2000, 8, 104-111.
7
(9)
Bouwer, E. J. In Handbook of Bioremediation; RD Norris, Hinchee, RE, Brown, R,
McCarty, PL, Semprini, L, Wilson, DH, Kampbell, M, Reinhard, EG, Bouwer, R,
Borden, C, Vogel, TM, Thomas, JM, Ward, CH, Eds.; Lewis Publishers: Boca Raton,
1994; pp 149-175.
(10)
Abelson, P. H. Science 1990, 250, 733.
(11)
Travis, C. C.; Doty, C. B. Environ. Sci. Technol. 1990, 24, 1464-1466.
(12)
Vogel, T. M.; Criddle, C. S.; McCarty, P. L. Environ. Sci. Technol. 1987, 21, 722736.
(13)
Hollinger, C.; Wohlfarth, G.; Diekert, G. FEMS Microbiol. Rev. 1998, 22, 383-398.
(14)
Chapelle, F. H. Ground-Water Microbiology and Geochemistry; John Wiley and
Sons, Inc.: New York, 1993.
(15)
McCarty, P. L. Science 1997, 276, 1521-1522.
(16)
Arp, D. J.; Yeager, C. M.; Hyman, M. R. Biodegradation 2001, 12, 81-103.
(17)
Mars, A. E.; Houwing, J.; Dolfing, J.; Janssen, D. B. App!. Environ. Microbiol. 1996,
62, 886-891.
(18)
Landa, A.; Sipkema, E.; Weijma, J.; Beenackers, A.; Dolfing, J.; Janssen, D. Appi.
Environ. Microbiol. 1994, 60, 3368-3374.
(19)
Mohn, W. W.; Tiedje, J. M. Microbiol. Rev. 1992, 56, 482-507.
(20)
Neumann, A.; Scholz-Muramatsu, H.; Diekert, G. Arch. Microbiol. 1994, 162, 295301.
(21)
Maymó-Gatell, X.; Chien, Y.; Gossett, J. M.; Zinder, S. H. Science 1997, 276, 1568-
(22)
Miller, E.; Wohlfarth, G.; Diekert, G. Arch. Microbiol. 1997, 168, 513-519.
(23)
Holliger, C.; Hahn, D.; Harmsen, H.; Ludwig, W.; Schumacher, W.; Tindall, B.;
Vazquez, F.; Weiss, N.; Zehnder, A. Arch. Microbiol. 1998, 169, 3 13-321.
(24)
Fantroussi, S. E.; Naveau, H.; Agathos, S. N. Biotechnol. Frog. 1998, 14, 167-188.
(25)
de Bruin, W.; Kotterman, M.; Posthumus, M.; Schraa, G.; Zehnder, A. App!. Environ.
Microbiol. 1992, 58, 1996-2000.
(26)
Maymó-Gatell, X.; Tandoi, V.; Gossett, J.; Zinder, S. App!. Environ. Microbiol. 1995,
64, 3928-3933.
1571.
8
(27)
National Research Council In Situ Bioremediation: When Does It Work?; National
Academy Press: Washington, D.C., 1993.
(28)
U.S. EPA 'Technical Protocol for Evaluating Natural Attentuation of Chlorinated
Solvents in Ground Water," 1998 EPA/600/R-98/128.
(29)
Lee, M. D.; Odom, J. M.; Buchanan, R. J. J. Ann. Rev. Microbiol. 1998, 52, 423-452.
(30)
Bekins, B.; Rittmann, B. E.; MacDonald, J. A. Eos 2001, 82, 57-58.
(31)
Lee, M. D.; Quinton, G. E.; Beeman, R. E.; Biehle, A. A.; Liddle, R. L.; Ellis, D. E.;
Jr., B. R. J. I md. Microbiol. Biotechnol. 1997, 18, 106-115.
(32)
Chapelle, F. H.; Lovely, D. R. App!. Environ. Microbiol. 1990, 56, 1865-1874.
(33)
Washington, J. W.; Cameron, B. A. Environ. Toxicol. Chem. 2001, 20, 1909-19 15.
(34)
Madsen, E. L. Environ. Sci. Technol. 1991, 25, 1662-1673.
(35)
Fetter, C. W. Applied Hydrogeology; Prentice Hall: Upper Saddle River, New Jersey,
(36)
McAllister, P. M.; Chiang, C. Y. Ground Water Monit. Rev. 1994, 14, 16 1-173.
(37)
Buscheck, T. E.; Alcantar, C. M. In Intrinsic Bioremediation; Robert E. Hinchee,
Wilson, John T., Downey, Douglas C., Eds.; Battelle Press: Columbus, 1995; pp 109-
1994.
116.
(38)
McNab, W. W. J.; Dooher, B. P. Ground Water 1998, 36, 983-987.
(39)
Istok, J. D.; Humphrey, M. D.; Schroth, M. H.; Hyman, M. R.; O'Reilly, K. T. Ground
Water 1997, 35, 619-63 1.
CHAPTER 2. IN SITU ANAEROBIC TRANSFORMATION OF
TRICHLOROFLUOROETHENE IN TRICHLOROETHENE-CONTAMINATED
tI11JiA7iUfl
Kimberly J. Hagemana*, Jonathan D. Jstokb, Jennifer A. Fieldc,
Timothy E. Buscheckd, and Lewis Semprinib
a
b
Department of Chemistry
Department of Civil, Construction, and Environmental Engineering
Department of Environmental and Molecular Toxicology
Oregon State University, Corvallis, Oregon 97331
d
Chevron Research and Technology Company
100 Chevron Way, Richmond, California 94802
Reproduced with permission from Environmental. Science and Technology 2001, 35, 1729173 5.
Copyright 2001 American Chemical Society.
III]
ABSTRACT
Methods are needed to obtain in situ information on the transformation rates of
trichloroethene (TCE), the most commonly detected organic groundwater contaminant. The
objective of this research was to investigate the potential for determining TCE transformation
rates in groundwater by measuring the transformation rate of its fluorinated surrogate,
trichiorofluoroethene (TCFE). To explore this hypothesis, the in situ transport behavior,
transformation pathway, and transformation rate of injected TCFE were determined in TCE-
contaminated groundwater using single-well, push-pull tests. Although transport behavior
varied between wells, TCFE, dichiorofluoroethene (DCFE), and TCE were transported
similarly to each other. In the shallow water-bearing zone, TCFE was reductively
dechlorinated to cis-DCFE, trans-DCFE, and (E)-1-chloro-2-fluoroethene (CFE), while co-
injected TCE was concurrently transformed to cis-dichloroethene (DCE), trans-DCE, 1,1
DCE, and a trace amount of chloroethene (CE). With added formate and the injected TCFE
concentration being a factor of 20 higher than that of TCE, the TCFE transformation rate
ranged from 0.053 to 0.30 jimol/L.-day, while that of TCE ranged from 0.009 to 0.0 12
.tmol/L-day.
Without added formate, the TCFE transformation rate decreased to 0.036
jirnol/L-day. In the deeper water-bearing zone, TCFE transformation occurred only after a lag
time of 55 days with added formate. No TCFE transformation occurred in groundwater that
had not previously exposed to TCE. The potential applicability for TCFE as an in situ
transport and transformation surrogate for TCE was demonstrated.
11
INTRODUCTION
Trichioroethene (TCE), a non-flammable solvent used in large quantities by industry,
is the most common organic groundwater contaminant (1) and is classified as a "probable
human carcinogen" (2). Due to evidence that subsurface microorganisms are capable of
degrading TCE under specific biogeochemical conditions, in situ bioremediation of TCEcontaminated groundwater is being investigated (3). TCE degradation under ,methanogenic
and sulfate-reducing conditions in laboratory (4) and field studies (5-10) has been reported.
Anaerobic TCE degradation occurs by reductive dechlorination, a reaction in which hydrogen
atoms sequentially replace chlorine substituents. Thus, in the commonly observed ICE
transformation pathway, TCE is sequentially reduced to the dichloroethene (DCE) isomers,
chloroethene (CE), and ethene (Figure 2.la).
In situ TCE transformation rates, which are needed to assess the potential for intrinsic
bioremediation and to design and monitor engineered bioremediation projects, have been
reported (11). However, the common method for estimating in situ rates, monitoring temporal
a)
,c
C1\
b)
,ci
/CC\
Cl\
/CC\
F
TCE
/0
/CC\
H\
Cl
H
trans-DCE
T
C\
TCFE
,C1
H
H
cis-DCE
Cl
/CC\
H
H
H
H\
Cl
/CC\
/
Cl
Cl\
,Cl
F
H
F
trans-DCFE
cis-DCFE
H
H
H
Cl
F
1CI, iF ethene
CE
H\
Cl
,H
/CC\
1,1-DCE
,H
H
CI\
/CC\
/
ethene
H
H
H
F
Cl
Cl\
Cl
x
,H
/CC\
F
1,1-DCFE
CI
H
H
F
(Z) 1CI, 2F ethene (E) 1CI. 2F ethene
(CFE)
H
,H
H
F
IF ethene (FE)
Figure 2.1. Reductive dechlorination pathways for (a) TCE (5) and (b) TCFE (12). The
predominant isomers and pathways are indicated by underlines and heavy arrows.
12
and spatial changes in TCE and transformation product concentrations, is problematic.
Vancheeswaran et al. (12) argued that transformation rates obtained in this way are often
ambiguous because (a) small changes in TCE and transformation product concentrations are
difficult to measure in the presence of high background concentrations, (b) microbiallygenerated transformation products cannot be distinguished from those that are present in the
background, and (c) concentration changes due to transformation are obscured by nonbiological processes such as advection, dispersion, sorptionldesorption and the dissolution of
non-aqueous phase TCE. Furthermore, groundwater tracer tests that involve the addition of
TCE to re-injected groundwater is problematic even if the endogenous TCE and its
degradation products are removed first (e.g., by air sparging). In these types of tracer tests,
dilution of the tracer test solution with background groundwater renders it impossible to
distinguish between injected and background TCE and its degradation products.
An alternative approach, in which the specified problems are avoided, involves
measuring the transformation rate of injected trichlorofluoroethene (TCFE) in TCEcontaminated groundwater and estimating the TCE transformation rate from that of TCFE
(12). In groundwater microcosm experiments, trichlorofluoroethene (TCFE) was reductively
dechlorinated to "fluorine-labeled" transformation products by a pathway analogous to that of
TCE (12) (Figure 2.lb). Moreover, with comparable initial TCFE and TCE concentrations,
zero-order TCFE and TCE transformation rates were similar in single-compound tests as well
as in tests where TCFE and TCE were present together. In free enzyme, corrinoid-mediated
experiments (13) with comparable initial TCFE and TCE concentrations, reductive
dechlorination of TCFE and TCE followed second-order kinetics and TCFE transformation
rates were 12 to 25 times higher than those of TCE.
The objective of this research was to determine the transport behavior, transformation
pathway, and transformation rate of TCFE under defined conditions in TCE-contaminated
groundwater at a former chemical manufacturing plant in the San Francisco Bay area. To this
end, single-well, push-pull tests with TCFE were conducted in two water-bearing zones with
different contaminant and biogeochemical characteristics. In a "push-pull" test, a prepared
test solution containing the compounds of interest and a conservative tracer is injected
("pushed") into the saturated zone of an aquifer and then extracted ("pulled") from the same
location (14, 15).
Breakthrough curves, which are used to assess the transport or
transformation behavior of the injected compounds, are constructed from samples collected
during the extraction phase of the test.
EXPERIMENTAL SECTION
Chemicals
Trichloroethene (TCE) (99.5% purity), chloroethene (CE) (97%), sodium formate, and
sodium bromide were obtained from Fisher Scientific (Fair Lawn, NJ). Cis-dichloroethene
(cis-DCE) (97%), trans-dichloroethene (trans-DCE) (98%), and 1,1 -dichloroethene (1,1 -DCE)
(99%)
were
obtained
from
Aldrich
Chemical
Company
(Milwaukee,
WI).
Trichlorofluoroethene (TCFE) (97% pure, containing 0.1% cis-dichloroethene (DCFE) and
0.1% trans-DCFE) and DCFE (98% pure mixture consisting of 50% cis and 50% trans
isomers)
were
obtained
from ABCR Chemicals (Karlsruhe,
Germany).
1,2-
chlorofluoroethene (97% pure mixture consisting of 31% (E) and 69% (Z) isomers) was
obtained from SynQuest Laboratories, Inc. (Alachua, FL). Fluoroethene (FE) (98%) was
obtained from Lancaster Synthesis (Pelham, NH). Ethene (19.2 ppm in nitrogen) was obtained
from Airco Special Gases (Vancouver, WA).
1-chioropropane and 1-chiorobutane, which
were used as internal standards for gas chromatography (GC) quantitation, were obtained from
Matheson Company (Cincinnati, OH) and Mallincrodt, Inc. (St. Louis, MO), respectively.
Site Description
The tests in this study were conducted in TCE-contaminated groundwater at a former
chemical manufacturing plant in the San Francisco Bay area where TCE reductive
dechlorination has been monitored in recent years (7, 8). Tests were conducted in two distinct
water-bearing zones, the A-zone and the C-zone. The A-zone is an unconfined shallow layer
composed mainly of placed fill over Bay Mud. The water table lies within 3 meters of the
ground surface. The groundwater velocity ranges from 1.5 to 6 meters per year. The C-zone
underlies the Bay Mud and is characterized by alluvial fan deposits, approximately 6 to 23
meters below the ground surface. Groundwater velocities range from 6 to 31 meters per year.
The water table slopes to the west in both zones.
Monitoring well installations and subsurface investigations began at the site in the
early 1980s. TCE and tetrachioroethene (PCE), pesticides, BTEX (benzene, toluene,
ethylbenzene, and xylene), and metals were detected in the A-zone. TCE and PCE were
detected in the C-zone. Neither DCE isomers nor CE were ever used or produced at the
facility.
Using a reductive dechlorination screening process that compares measured
14
concentrations of contaminants and biogeochemical indicators to threshold values (16),
Buscheck found that there was strong evidence for reductive dechlorination in the A-zone and
weaker evidence for reductive dechlorination in the C-zone (8).
Push-Pull Tests
A series of push-pull tests was conducted to obtain information on aqueous TCFE and
TCE transport and transformation in the selected A-zone well, RI-bA, and in selected C-zone
wells, GW-15C and GW-21C. These wells contain a range of background contaminant and
Table 2.1. Contaminants and Biogeochemical Indicators in Selected Wells
Concentration (,lmol/L)a
A-zone well
C-zone wells
R1-1OA
GW-15C
GW-21C
trichloroethene (TCE)
NDb
240
ND
cis-1,2dichloroethene
(cis-DCE)
0.0 17
ND
ND
benzene
0.078
0.32
ND
toluene
0.0041
ND
ND
ethylbenzene
0.0029
ND
ND
total xylenes
0.005
ND
ND
ethene
ND
0.0043
ND
ethane
ND
ND
ND
18
0.26
ND
15,000
ND
ND
dissolved oxygen
130
5.6
18
nitrate-N
ND
ND
ND
sulfate
960
490
170
total dissolved iron
95
ND
ND
methane
total organic carbon
aSamples collected in May-June, 1999. Chlorinated hydrocarbons and BTEX compounds by
EPA method 8021B; ethene, ethane, and methane by RSK-175; total organic carbon by EPA
9060; dissolved oxygen by membrane electrode probe; nitrate-N and sulfate by EPA 9056;
iron by EPA 60 lOB. bNOt detected.
15
biogeochemical indicator concentrations (Table 2.1).
Test solutions consisted of tap water, bromide (to serve as a
Transport Tests.
conservative tracer), TCFE and in one case, TCE (Table 2.2). Although it would have been
desirable to use site groundwater in these experiments, the use of tap water was required to
obtain regulatory approval to conduct these tests at this site. Note that because the purchased
TCFE standard contained 0.1% cis-DCFE and 0.1% trans-DCFE, -0.015 jtmol/L of these
compounds were also injected in every test. The test solution was prepared by adding
bromide to the tap water and then sparging the solution for at least 4 hours with compressed
air to mix and aerate the solution prior to injection. A concentrated aqueous solution of TCFE
(and TCE, where applicable) was stored in a collapsible metallized-film gas-sampling bag
(Chromatography Research Supplies, Addison, IL) to prevent volatilization losses during
injection (Figure 2.2). TCFE and TCE were added to the tap water/bromide solution by
metering the solution from the bag into the main injection line with a piston pump (Fluid
Metering Inc., Oyster Bay, NY) (Figure 2.2). TCFE/TCE equilibration between the inner
polyethylene layer of the bag and the TCFE/TCE solution was established by waiting at least
2 hours between filling the bag and starting the injection. TCFE/TCE equilibration between
Table 2.2. Push-Pull Test Descriptions
Test Solution Composition
bromide
TCFE
TCE
formate
zone
(mmol/L)
(jtmolIL)
(p.moVL)
(mmollL)
RI-bA
A
1.3
8.9
0.019
--
2
GW-21C
C
1.2
15
3
GW-15C
C
1.3
14
0.78
2.0
--
8.3
test
well
Transport Tests
Transformation Tests
4
RI-bA
A
1.2
16
5
RI-bOA
A
1.3
13
6
GW-15C
C
1.3
19
7
GW-21C
C
1.4
33
V
the injection line tubing and the test solution was established by purging the injection lines
with test solution for 10 minutes prior to starting the injection phase. Equilibration times were
determined in preliminary laboratory experiments.
The injection/extraction procedure for tests in the A- and C-zones were not identical
since well diameters were 2.5 cm in the A-zone and 10 cm in the C-zone. In the A-zone test,
50 L of test solution were injected into the bottom of the well at a flow rate of 0.2 L/min
(Figure 2.2a).
In C-zone tests, -250 L of test solution were injected between a pair of
inflatable packers at a rate of -2 L/min (Figure 2.2b). The packers were used to isolate a
meter-long section of the well screen. In all tests, the test solution was injected through 6-mm
nylon-braided tubing (Kuiyama Co., Santa Fe Springs, CA) into the well with a Masterfiex
peristaltic pump (Barnant Co., Barrington, IL). Five to ten samples of the test solution, which
Compressed Air or Argon
11
Carboy with
fI111Lest Solution
Per.staltic
Sampling
Valve
Piston
Pump
\
Quick
Connect
Collapsible Bag Containing
TCFE/TCE
Sampling
ii Valve
1
Compressed
Air
Land Surface
Land Surface
Water Table
Water Table
Upper Packer
Injector
Well Casing
Well Casing
(a)
I
Lower Packer
(b)
Figure 2.2. Experimental set-up for test solution injection in (a) the A-zone well and (b)
C-zone wells (not drawn to scale).
17
were analyzed to determine injected concentrations, were collected from the sample valve
during injection.
Immediately after completion of the injection phase, the carboy and the section of the
injection line between the two quick connects were removed and the flow direction of the
peristaltic pump was reversed. In the A-zone test, 67 L of test solution/groundwater mixture
were extracted at a rate of -0.2 L/min; 20 samples were collected. In C-zone tests, 500 L of
test solution/groundwater mixture were extracted at a rate of 2 L/min; 50 samples were
collected. All samples were collected in volatile organic analysis vials without headspace,
shipped on ice, and stored at 4 °C until analysis. Samples for volatiles were collected in
duplicate and duplicate analyses were performed on approximately 10% of the samples.
Samples for bromide were collected and analyzed in duplicate. All samples except those for
bromide were preserved in 0.75% (v/v) concentrated HC1.
Transformation
Tests. Two transformation tests were conducted in the A-zone well
and one was conducted in each C-zone well. Test solutions consisted of tap water, bromide,
TCFE, and in some cases, ICE and formate (Table 2.2). The test solution was prepared by
adding bromide and formate to the tap water and then sparging the solution for at least 4 hours
with compressed argon to mix and remove dissolved oxygen prior to the start of the injection
phase. TCFE and TCE were then added to the test solution using a piston pump as described
for the transport tests. The injected test solution volumes and injection flow rates for the
transformation tests were identical to those described for the transport tests. Samples of the
test solution/groundwater mixture were collected approximately once per week for up to 82
days. Prior to sample collection, A- and C-zone wells were purged by extracting 0.3 and 12 L
of groundwater, respectively.
Samples were preserved and stored as described for the
transport tests.
Analytical Methods
Concentrations of TCFE, DCFE, CFE, TCE, DCE, and CE were determined by
headspace analysis with gas chromatography/mass spectrometry (GC/MS).
Qualitative
analysis of FE and ethene were performed by solid-phase microextraction with GC/MS. The
GC/MS system was composed of a Hewlett-Packard (Palo Alto, CA) model 5890 GC and
5972 series MS detector. Chromatographic separations were performed on a Supelco
18
(Bellefonte, PA) 30 m
0.32 mm
4 p.m SPD-1 column.
The identities of the TCFE
degradation products were confirmed by comparing their spectra, which were obtained by
x
x
operating the MS in scan mode, to published spectra (12). The MS was operated in selected
ion monitoring mode for quantitation. 1-chioropropane and 1-chlorobutane were used as
internal standards.
The quantitation limits (signal/noise = 10) were
0.005 p.mol/L for
analysis by headspace and -0.2 imol/L for analysis by solid-phase microextraction.
Bromide concentrations were determined by external calibration using a Dionex
(Sunnyvale, CA) model DX-120 ion chromatograph equipped with an electrical conductivity
detector and a Dionex AS14 column. Formate concentrations, as well as those of its potential
degradation product, acetate, were determined by external calibration using a Waters Alliance
(Milford, MA) high pressure liquid chromatograph (HPLC). The HPLC was equipped with a
model 2690 separations module, a model 996 photodiode array detector, and a Phenomenex
(Torrance, CA) 150mm x 4.60mm x 5 p.m LunaCl8 column.
RESULTS & DISCUSSION
Transport Tests
Transport tests were conducted to determine the relative transport behavior of injected
TCFE, DCFE, and TCE in A- and C-zones prior to initiation of the transformation tests. The
potential for anaerobic transformation of injected TCFE and TCE during transport tests was
minimized by (a) air-saturating the test solution prior to injection and (b) completing each test
in less than 10 hours. Transport test data were interpreted using breakthrough curves that
display relative concentration against the cumulative volume extracted at the time the sample
was collected divided by the volume of injected test solution. Relative concentration is
defined as the measured concentration of a compound in a sample divided by the average
concentration of the same compound in the injected test solution.
In the transport test conducted in RI-bA (test 1), the extraction phase breakthrough
curves for TCFE and TCE were nearly identical (Figure 2.3a), indicating that TCFE and TCE
were transported similarly. However, deviation of the TCFE/TCE breakthrough curve from
that of bromide indicates that TCFE and TCE were not transported conservatively during the
test. Since the test was conducted in less than 10 hours, we assume that the observed mass
loss was not due to transformation but was instead caused by TCFE/TCE sorption to sediment
'9
1.
0.
0
::
0.
0.1
U.h
U.O
tJ.4
J.O
1.Z
1.13
1.4
Volume Extracted I Injected
1.0
b)
0.8
Bromide
0
v
7
TCFE
cis-DCFE
trans-DCFE
0.2
DX1
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Volume Extracted I Injected
Figure 2.3. Breakthrough curves indicating (a) non-conservative transport of TCFE and
TCE in A-zone well RI-1OA (test 1) and (b) conservative transport of TCFE and DCFE
isomers in C-zone well GW-2 IC (test 2).
20
organic matter and/or TCFE/TCE partitioning to nonaqueous phase liquids that may be
present in the A-zone. However, no NAPLs were detected during construction of RI-bA.
Mass balance calculations indicated that 58, 41, and 38% of injected bromide, TCFE, and
TCE were extracted during the test, respectively. The similarity in TCFE and TCE transport
behavior supports the hypothesis that TCFE can be used as a surrogate for TCE in TCEcontaminated groundwater.
The transport behavior of cis- and trans-DCFE could not be
determined in this test due to interference with residual cis- and trans-DCFE from a prior test
conducted in this well.
The effects of injected solute sorption on push-pull test extraction phase breakthrough
curves were studied in numerical simulations (17, 18). Based on those simulations, the TCFE
and TCE transport behavior observed in test 1 cannot be attributed to equilibrium sorption
with either linear or Langmuir isotherms.
Thus, a more complicated sorption process,
possibly influenced by diffusion-limited (nonequilibrium) mass transfer may be occurring.
In the transport test conducted in well GW-21C (test 2), TCFE, cis-DCFE, and trans-
DCFE breakthrough curves were nearly identical to that of bromide (Figure 2.3b), indicating
that all three compounds were conservatively transported. Mass balance calculations indicated
that 69, 77, 64, and 70% of injected bromide, cis-DCFE, trans-DCFE, and TCFE were
extracted during the test, respectively. TCFE was also conservatively transported in well GW-
I SC (test 3) (data not shown). The variation in transport behavior between wells may be due
to varying soil organic carbon content or to the potential presence of NAPLs within the zone
of influence of injected test solutions in some wells.. However, when the complete series of
transport tests, which included two tests in the A-zone and four in the C-zone, is considered,
no correlations between transport behavior and zone, background chlorinated ethene
concentrations, or total organic carbon could be identified (data not shown).
Transformation Tests
Transformation tests were conducted (a) to determine if injected TCFE could undergo
reductive dechlorination in the selected water-bearing zones, (b) to compare the in situ
transformation pathways and rates for TCFE and TCE, and (c) to determine if reductive
dechlorination of TCFE and TCE is limited by electron donor availability at this site.
A-Zone Tests.
Test 4 was designed to test the hypothesis that the addition of formate
would stimulate reductive dechlorination of TCFE and TCE in the A zone by serving directly
as an electron donor or by stimulating other microbial activity that produce electron donors
21
that could be used by dechlorinating organisms. Relative concentrations for the bromide
tracer (the measured bromide concentration C divided by the bromide concentration in the
injected test solution, C0) decreased with time as the test solution was gradually diluted with
site groundwater (Figure 2.4a). By the end of the test, the bromide relative concentration was
C/C0
= 0.39 indicating that this sample was a mixture of test solution (39 %) and groundwater
(61 %). The effects of dilution on the concentration of a conservatively transported compound
can be removed by dividing the compound's measured concentration by the relative
concentration of the co-injected tracer (19). Conservative transport of formate was assumed
for this study because of the high water solubility and negative charge of formate. Measured
formate concentrations were divided by
C/C0
to produce "dilution-adjusted" formate
concentrations (Figure 2.5a and b). The rapid decrease in formate concentration suggests that
an active anaerobic microbial community capable of utilizing formate was present.
However, acetate, a potential fermentation product of formate resulting from acetogenesis,
was not detected.
Measured TCFE concentration also decreased during test 4 (Figure 2.4b).
The
observed production of cis-DCFE, trans-DCFE, and (E)-1-chloro-2-fluoroethene (CFE)
indicates that reductive dechlorination of injected TCFE occurred during this test. To increase
the interpretability of the results and to compute transformation rates, it was necessary to
adjust measured concentrations for dilution. However, the method used to adjust measured
formate concentrations for dilution could not be employed since the transport test conducted in
this well indicated non-conservative transport of TCFE, which means that relative
concentrations of the bromide tracer cannot be used to correct for dilution of TCFE and its
transformation products.
Instead, an alternate dilution-adjustment method, which uses
concentration ratios (19), was devised.
The method assumes that: (a) the transport behaviors of TCFE and its transformation
products are identical, and (b) all potential TCFE transformation products are identified and
quantified in each sample.
The first assumption is supported by the observed identical
transport behavior for TCFE and DCFE isomers in tests conducted in the A- and C-zone tests
1 and 3 (Figure 2.3) and by computed organic matter-water partition constants
TCFE, cis-DCFE, trans-DCFE, CFE, and FE (log
respectively).
Values of
Korn
Korn
= 2.7,
1.9, 1.9,
(K0m) for
2.1, and 1.3,
were computed from octanol-water partition constants
(K0)
(20), which were estimated from structural group contributions (21). For example, assuming
22
a)
2.0
U
1.8
,
A
1.0
Formate
Bromide
0.9
1.6
0.8
1.4
0.7
0.6
E
0.5
0.4
0.3
0.4
0.2
0.2
0.1
0.0
0.0
10
0
30
20
40
50
60
70
Time (Days)
18
0
16
V
v
E
A
TCFE
cis-DCFE
trans-DCFE
CFE
(Sum C/Sum Co)
u.I
1.0
2.0
0.10
1.5
12
1.0
L
10
,
'
2
0.8
0.05
0.5
0.6
0.00
0.0
0.4
6
4
0.2
H
2
0
*;4:TT:i
10
20
30
40
50
0.0
60
70
Time (Days)
Figure 2.4. Test 4 data indicating (a) changes in concentration for formate and relative
concentration for bromide and (b) reductive dechlorination of injected TCFE to cis-DCFE,
trans-DCFE, and CFE in A-zone well RI-i OA. Measured concentrations are not adjusted
for dilution.
23
equilibrium linear sorption and estimated values for aquifer organic matter content in the A
and C zones (forn = 0.0006 0.03), bulk density (2.12 glcm3), and porosity (0.2), estimated
retardation factors for TCFE and its transformation products ranged from I to 26.
The
second assumption
is
considered valid because
all
currently known
transformation products of TCFE and TCE (Figure 2.1) were analyzed by GC/MS. Although
CO2 and CH4 production during reductive dechlorination of TCE in microcosm experiments
has been observed by others, their production is thought to be the result of a combination of
anaerobic chloroethene oxidation to acetate and acetotrophic methanogenesis (22-24). Since
chlorofluoroethene was not produced until day 25 of the test, production of CO2 and CH4 from
chlorofluoroethene is likely not responsible for the observed reduction in TCFE concentrations
in test 4.
Measured concentrations of TCFE and its transformation products were corrected for
dilution by dividing the measured concentration by the dilution factor (SumC/SumC()) (Figure
2.4b) defined as
SumC
[TCFEI + [cis DCFE] + [trans - DCFE] + [CFE]
[TCFE]0 + [cis DCFE](, + [trans
+ [CFE]()
where, for example, [TCFE] and
[TCFE]O
(I)
are the measured TCFE concentrations in a sample
and in the injected test solution, respectively.
The dilution corrected concentrations for TCFE and its transformation products during
test 4 are plotted in Figure 2.5a and for co-injected TCE and its transformation products
during the same test in Figure 2.5b. Dilution adjustments for TCE and its transformation
products were performed using an analogous equation to eq I. Note that dilution-adjusted
concentrations are displayed in Figure 2.5 and subsequent figures. The dilution-adjusted
concentrations show the overall similarities in transformation pathways and rates for TCFE
and TCE. TCFE transformation to cis-DCFE, trans-DCFE, and CFE occurred concomitantly
with the transformation of TCE to cis-DCE, trans-DCE, and 1,1-DCE (Figure 2.5). The
observed similarity in the in situ TCFE and TCE transformation pathways, including isomer
predominance (Figure 2.1), indicates that it may be possible to use TCFE to determine the
extent of TCE transformation, an important parameter in bioremediation design. Slower
dechlorination between 0-18 days and faster transformation between 18-67 days, were similar
for TCFE and TCE, which suggests that TCFE and TCE transformations were affected by
similar rate-controlling factors. The TCFE concentration decreased at a nearly linear rate,
24
16
E
U
'
14
v
12
H
2.0
a)
TCFE
CiL-DCFE
1.8
trans-DCFE
CFE
Forinate
1.6
1.4
10
1.2
2
1.01
0.8
2
0.6
U
0.4
0.2
U
0.0
0
10
20
30
40
50
60
70
Time (days)
0.8
v
7
TCE
2.0
b)
ci-DCE
1.8
trans-DCE
1,1-DCE
1.6
Formate
1.4
05
U
I..
L
1.2
0.4
1.01
0.8
0.3
0.6
0.2
0.4
0.1
U
0.2
0.0
0.0
0
10
20
30
40
50
60
70
Time (days)
Figure 2.5. Reductive dechlorination of (a) injected TCFE to cis-DCFE, trans-DCFE,
and CFE and (b) TCE to cis-DCE, trans-DCE, and 1,1-DCE in A-zone well RI-bA with
added formate as an exogenous electron donor (test 4). Measured concentrations are
adjusted for dilution.
25
16
14
.<
12
10
V
v
TCFE
cis-DCFE
trans-DCFE
CFE
4
___,
2
11
0
30
20
10
50
40
60
80
70
Time (days)
Figure 2.6. Reductive dechlorination of injected TCFE to cis-DCFE, trans-DCFE and
CFE in A-zone well RI-i OA without added formate as an exogenous electron donor (test
5). Measured concentrations are adjusted for dilution.
20
E
9
y
I
7
y -___
y
8
18
7
16
C-)
C-
14
12
\
JO
CC-
rD
V
\
8
6i0
TCFE
cis-DCFE
(rans-DCFE
V
V
\
\
3
6
2
0
2
4
1
2
0_
fi
0
10
20
30
40
0
50
60
70
80
Time (days)
Figure 2.7. Reductive dechlorination of injected TCFE to cis- and trans-DCFE occurring
after 55 days in C-zone well GW-15C with added formate as an exogenous electron donor
(test 6). Measured concentrations are adjusted for dilution.
c1
Table 2.3. Reductive Dechlorination Rates for TCFE, TCE, and Their Transformation
Products in Push-Pull Transformation Tests
Transformation Rates .tmoI/L.day)*
A-zone tests
Test 4: with formate
C-zone tests
Test 5: without formate
Test 6: with formate
Test 7: without formate
days
0-18
18-67
0-80
0-55
55-82
0-80
TCFE
-0.053
-0.3
-0.036
0
-0.21
0
cis-DCFE
+0.044
+0.19
+0.029
0
+0.20
0
trans-DCFE
+0.009
+0.011
+0.0049
0
+0.0 1 1
0
CFE
0
+0.098
+0.0025
0
0
0
TCE
-0.009
-0.012
cis-DCE
+0.009
+0.009
trans-DCE
+0.00039
+0.00072
1,1-DCE
+0.00015
+0.0022
* "+" indicates increasing concentrations; 'i" indicates decreasing concentratkns
27
36
'V
V
V _ V
V
V
_
'V
32
28
v
v
24
TCFE
cis-DCFE
lrans-DCFE
20
16
12
8
4
0
I
0
10
I
20
30
40
I
I
50
60
70
80
Time (days)
Figure 2.8. Absence of reductive dechlorination of injected TCFE in C-zone well GW21 C without added formate as an exogenous electron donor in groundwater that had not
previously been exposed to TCE or other co-contaminants (test 7). Measured
concentrations are adjusted for dilution.
which was determined by linear regression analysis to be 0.053 tmol-L between 0-18 days
and 0.30 imol/L-day between 18-67 days (Table 2.3). Ninety-three percent of the injected
TCFE was transformed during the 67-day test. The TCE concentration decreased in a nearly
linear manner at a rate of 0.009 jimol/L-day between 0-18 days and 0.0 12 jtmol/L-day
between 18-67 days (Table 2.3). Ninety-seven percent of the injected TCE was transformed
to identified products during the test.
Although the overall percentages of TCFE and TCE transformation were similar;
TCFE was transformed at a rate 5.8 times larger than TCE between 0-18 days and 25 times
larger than TCE between 18-67 days. This rate difference is likely due to the higher injected
TCFE concentration (20 times higher than TCE). TCE was injected at a lower concentration
in this test to meet regulatory requirements.
28
Test 5 was designed to determine if TCFE transformation rates were limited by
electron donor availability by conducting a second test in well RI-bA but without the
addition of formate (Table 2.2). TCFE was again transformed to cis-DCFE, trans-DCFE, and
CFE; FE was not detected (Figure 2.6). TCFE concentrations decreased and cis-DCFE, trans-
DCFE, and CFE concentrations increased linearly. The TCFE transformation rate was 0.036
.tmol/L-day and 19% of the injected TCFE was transformed during the 80-day test. These
results indicate that TCFE transformation rates in well RI-bA were limited by the availability
of electron donors since the transformation rate in test 5 without formate (Figure 2.6) was 1.5
to 8.3 times smaller than that in test 4 with formate (Figure 2.5).
This observation is
significant because it suggests that TCE transformation rates may be increased at this site by
supplying exogenous electron donors.
C-zone Tests.
Test 6, conducted in well GW-15C, was designed to determine the rate
of TCFE transformation in the C-zone with formate added as an electron donor. Formate was
utilized at a slower rate in this test compared to test 4 conducted in the A-zone, and was not
completely degraded until day 27 (Figure 2.7). Acetate was not detected. No reductive
dechlorination of TCFE was observed before day 55 (Figure 2.7). TCFE concentrations
decreased at a rate of 0.21 jimol/L-day between 55-82 days (Table 2.3). Thirty-two percent of
the injected TCFE was transformed to cis-DCFE and trans-DCFE during the 80-day test; CFE
and FE were not detected. It was not possible within the scope of this project to conduct a
second test in well GW-15C without formate.
Test 7, conducted in well GW-21C, was designed to determine if reductive
dechlorination would occur in groundwater that had not previously been exposed to TCE or
other co-contaminants (Table 2.1). Transformation of TCFE was not observed in well GW21C (Figure 2.8), which is consistent with the hypothesis that dechlorinating microorganisms,
if present at this location, would not be active in an uncontaminated portion of the aquifer. The
observed slower rate of TCFE transformation in tests conducted in the C-zone compared to
tests conducted in the A-zone is consistent with the conclusions drawn from the reductive
dechlorination screening process conducted at this site by Buscheck (8).
Although
transformation was not observed in this well, TCFE was recovered during the 3-month long
field experiment, which indicates that the push-pull test format is appropriate for this type of
field application.
29
In summary, the potential applicability of TCFE as an in situ surrogate for ICE
transport and transformation was demonstrated by the similarity in TCFE and TCE transport
behavior, the composition of reductive dechlorination transformation products that formed in
situ, and in the general agreement between the rates of transformation for TCFE and TCE.
ACKNOWLEDGEMENTS
We wish to thank Mark Humphrey, Ralph Reed, and Andrew Lundmark for analytical
assistance and Brian Davis, Kevin Tam, and Angelito Tirona for field assistance. Supelco,
Inc. (Bellefontaine, PA) provided the GC column. Funding was provided by the U.S. EPA-
sponsored Western Region Hazardous Substance Research Center and by Chevron
Environmental Management Company.
LITERATURE CITED
(1)
(2)
Domenico, P. A.; Schwartz, F. W. Physical and Chemical Hydrogeology; John Wiley
and Sons: New York, 1990.
"Public Health Goal Document for Trichloroethylene"; California EPA Office of
Environmental Health Hazard Assessment, 2000.
(3)
National Research Council. In Situ Bioremediation: When Does It Work?; National
Academy Press: Washington, D.C., 1993.
(4)
Vogel, T. M.; Criddle, C. S.; McCarty, P. L. Environ. Sci. Technol. 1987, 21, 722.
(5)
Semprini, L.; Kitandis, P. K.; Kampbell, D. H.; Wilson, J. T. Water Resour. Res.
(6)
1995, 31, 1051.
Major, D. W.; Hodgins, E. W.; Butler, B. J. In On-Site Bioreclamation: Processes for
Xenobiotic and Hydrocarbon Treatment; Hinchee, R. E., Olfenbuttel, R. F., Eds.;
Butterworth-Heinemann: Boston, 1991.
(7)
Buscheck, T. E.; O'Reilly, K. T.; Hickman, G. In In Situ and On-Site Bioremediation
(3); Battelle Press: Columbus, OH, 1997; Vol. 3, pp 149.
(8)
Buscheck, T. E. ; Chevron Research and Technology Company, unpublished report,
(9)
Kaster, M. Appi. Environ. Microbiol. 1991, 57, 2039.
(10)
McCarty, P. L.; Wilson, J. T. In Bioremediation of Hazardous Wastes; US EPA
Office of Research and Development, EPA., Washington, D.C., 1992, pp 68;
1998.
EPA/600/R-92/1 26.
(11)
Suarez, M. P.; Rifai, H. S. Biorem. J 1999, 3, 337.
(12)
Vancheeswaran, S.; Hyman, M. R.; Semprini, L. Environ. Sci. Technol. 1999, 33,
2040.
30
(13)
Glod, G.; Angst, W.; Hollinger, C.; Schwarzenbach, R. P. Environ. Sci. Technol.
1997,
31, 253.
(14)
Istok, J. D.; Humphrey, M. D.; Schroth, M. H.; Hyman, M. R.; O'Reilly, K. T. Ground
Water 1997, 35, 619.
(15)
Schroth, M. H.; Istok, J. D.; Conner, G. T.; Hyman, M. R.; Haggerty, R.; O'Reilly, K.
T. Ground Water 1998, 36, 924.
(16)
Wiedemeier, T. H.; Swanson, M. A.; Moutoux, D. E.; Wilson, J. T.; Kampbell, D. H.;
Hansen, J. E.; Haas, P. In Symposium on Nat ural Attenuation of Chlorinated Organics
in Ground Water: Dallas, TX, 1996, pp 35; EPA/540/R-96/509.
(17)
Schroth, M. H.; Istok, J. D.; Haggerty, R. Adv. Water Res. 2000, 24, 105.
(18)
Istok, J. D.; Field, J. A.; Schroth, M. H.; Sawyer, T. E.; Humphrey, M. D. Ground
(19)
Madsen, E. L. Environ. Sci. Technol.
(20)
Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental Organic
(21)
Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis in Chemistiy and
Biology; John Wiley and Sons, Inc.: New York, 1979.
(22)
Bradley, P. M.; Chapelle, F. H. Environ. Sci. Technol.
1999,
33, 3473.
(23)
Bradley, P. M.; Chapelle, F. H. Environ. Sci. Technol.
1999,
33, 653.
(24)
Bradley, P. M.; Chapelle, F. H. Environ. Sci. Technol. 2000, 34, 2761.
Water 1999, 37, 589.
1991,
25, 1662
Chemistry; John Wiley and Sons, Inc.: New York, 1993.
31
CHAPTER 3. "FORCED MASS BALANCE" TECHNIQUE FOR ESTIMATING IN
SITU TRANSFORMATION RATES OF SORBING SOLUTES IN GROUNDWATER
Kimberly J. Hageman*', Jennifer A. Fieldb, Jonathan D. Istokc, Martin H. Schrothd
aDepartment of Chemistry, bDepartment of Environmental and Molecular Toxicology,
cDepartment of Civil, Construction, and Environmental Engineering,
Oregon State University, Corvallis, OR 97331-7301
dlnstitute of Terrestrial Ecology
Swiss Federal Institute of Technology (ETH), Zurich, Switzerland
Reproduced with permission from Environmental Science and Technology, submitted for
publication. Unpublished work copyright 2003 American Chemical Society.
32
ABSTRACT
A method for estimating in situ transformation rates of sorbing solutes in groundwater
is presented. The method utilizes a novel data processing technique called "forced mass
balance" (FMB) to remove the effects of transport processes from reactant concentrations
measured during single-well, "push-pull" tests. The effectiveness of the FMB technique was
evaluated by quantifying errors in rates derived by applying FMB to push-pull test data
generated by a numerical model. Results from simulated tests indicated that errors in derived
rates increase as the test duration, groundwater velocity, and ratio of reactant to product
retardation factors increase. In addition, errors in derived rates increase as the reaction rate
constant and aquifer dispersivity decrease. As a demonstration, the FMB technique was used
to derive an in situ reductive dechlorination rate for trichlorofluoroethene (TCFE) using data
from a field push-pull test. Error analyses indicated that the in situ TCFE transformation rate
was underestimated by a factor of 1.1 to 2. Thus, the FMB technique makes it possible to
estimate in situ transformation rates of sorbing solutes and when FMB is coupled with
computer modeling, errors in derived in situ rates can be quantified.
33
INTRODUCTION
Protocols for intrinsic and engineered approaches to groundwater remediation include
steps for determining in situ transformation rates. However, in situ transformation rates are
difficult to determine because reactive solute (reactant) concentrations in groundwater are
affected by a combination of transformation and transport (advection, dispersion, sorption)
processes.
A number of methods use spatial differences
in
reactant and product
concentrations to determine in situ transformation rates (1-5).
Spatial differences are
determined by collecting data from multiple wells arranged in transects or grids. However,
wells are expensive and time-consuming to construct and some of these methods are limited to
use with steady-state solute plumes
serve as a tracer (3,
(3, 5, 6)
or plumes containing a nonreactive solute that can
4).
Temporal differences in reactant and product concentrations measured at a single well
can also be used to determine in situ transformation rates.
Single-well methods are
advantageous in that only one well is needed per test, they are not limited to plumes with
special characteristics, and site-scale variability can be assessed by comparing results from
different wells at a single site. One approach is to measure reactant concentrations at a single
well over time and then use site-specific hydrologic properties in combination with "full
inverse" modeling to differentiate changes in reactant concentrations due to transformation
from those due to transport processes (7). An alternative approach involves determining in
situ transformation rates from data collected during single-well "push-pull" tests. Push-pull
tests are conducted by injecting ("pushing") an aqueous test solution containing a nonsorbing,
nonreactive tracer and one or more reactants into an aquifer via a monitoring well (8).
Samples of the test solution/groundwater mixture are then extracted ("pulled") from the same
well over time and analyzed for tracer, reactant, and product concentrations.
The in situ transformation rate of the reactant is then determined using a data
processing technique to remove the effects of transport processes from measured
concentrations of the reactant.
In situ transformation rates were determined for injected
reactants, including oxygen (9), nitrate (9, 10), sulfate (11, 12), acetate (10), and formate (13),
from push-pull test data using a data processing technique that is hereafter referred to as
"tracer-normalization."
With this technique, the concentration of a nonsorbing reactant
measured in an extraction sample is adjusted by dividing it by the relative concentration of the
co-injected tracer measured in the same sample
(14, 15).
Thus,
34
[A]
(1)
[T]
where [AITN is the tracer normalized concentration of the nonsorbing solute, A; [A] and [TJ are
measured concentrations of A and the co-injected tracer, T, respectively; and [fl0 is the
injected concentration of T. Since the concentration of the tracer is affected by transport but
not by transformation, the effect of transport processes on measured concentrations is
expressed quantitatively by
[fl/[T]0.
The validity of the tracer-normalization technique was
verified by simulating push-pull tests with a modeling program and then comparing the rate
constants derived from tracer-normalized concentrations to the rate constants used as inputs in
the simulation program (14, 15).
However, the tracer-normalization technique was designed for use with nonsorbing
reactants only (14, 15). Based on the desire to determine in situ reductive dechlorination rates
of the sorbing reactant, trichlorofluoroethene (TCFE), Hageman et al. designed a data
processing technique to remove the effects of transport processes (including sorption) from
TCFE concentrations measured during push-pull tests (13).
TCFE, which is sequentially
transformed by reductive dechlorination to dichlorofluoroethene (DCFE), chiorofluoroethene
(CFE), and fluoroethene (FE) in anaerobic environments, is of interest because it can be used
as a surrogate for the priority pollutant, trichioroethene (TCE) (13, 16). The data processing
technique presented by Hageman et al. is based on the assumption that TCFE and all of its
products experience linear equilibrium partitioning and are transported identically to each
other, but differently from the tracer.
Adjusted TCFE concentrations, from which in situ TCFE transformation rates are
obtained, are calculated from
[TCFE] adjusted
[TCFE]
/
(2)
o
where [TCFEI is the measured aqueous concentration of TCFE in an extraction sample and
1a is an adjustment factor. The adjustment factor is calculated for each extraction sample
from
[TCFE]+[DCFE]+[CFE]+[FE]
where, for example, [DCFE] and
[DCFE]O
(3)
are the measured aqueous concentrations of DCFE
in an extraction sample and in the injected test solution, respectively. Since the sum of the
35
concentrations of TCFE and all of its products
is
affected by transport but not by
transformation, the effects of transport processes on measured concentrations are expressed
quantitatively by
Note that this data processing technique requires quantification of all
potential transformation products of TCFE. FE was not included in the equation defining
in Hageman et al. because it was not detected during field tests; however, it is included here
for completeness.
Adjusted concentrations for each transformation product are calculated
with equations analogous to eq 2. Adjusted product concentrations are used to assess product
distribution and the extent of dechlorination; however, transformation rates of intermediate
products are not readily determined since they are concurrently being produced and
transformed.
The objective of this work was to present and evaluate a data processing technique for
use in determining in situ transformation rates of sorbing reactants whose products are not
necessarily transported identically to the reactant or to each other. The technique, called
"forced mass balance" (FMB), differs from that previously presented by Hageman et al. (/3)
in that the effects of transport processes (including sorption) are removed from total (aqueous
plus sorbed) concentrations instead of just aqueous concentrations. Thus, for the reaction in
which A is sequentially transformed to B and C, FMB-adjusted concentrations of the sorbing
reactant, A, are calculated from
[A]
aq+s
where
Ajaq+s
(4)
'aq±s,o
is the total concentration of A associated with an extraction sample.
adjustment factor,
The
aq+.'aq+s,o, is calculated for each extraction from
/
where, for example,
[BIaq+S
and
aq+s.o
[A]aq±s +[B]aq+ +[CLqs
[A]aq±so + [B]aq+o + [CIaq±o
(5)
[B1aq+o are the total concentrations of B in an extraction
sample and in the injected test solution, respectively. The total concentration of A,
[AIaq+c, for
example, is calculated from
[A]aq+s
Aaq +A
Vaq+Vs
(6)
where
Aaq
is the mass of A in the extraction sample of volume, Vaq, and A., is the mass of A in
the volume of aquifer sediments, V, associated with the extraction sample based on the
aquifer's porosity, n. The sorbed mass of A, A, for example, is calculated from
A
where
Korn
c
[A1aq *K om * Jom
*
S
*
(7)
g
is the organic matter/water distribution constant, forn is the fraction of organic matter
in aquifer sediments, and
g
is the grain density of aquifer sediments.
FMB-adjusted
concentrations for each transformation product are calculated with analogous equations.
"Forced mass balance" is an illustrative term for this technique because the sum of FMBadjusted reactant and product concentrations is conserved. Note that the aqueous reactant and
product concentrations adjusted by the Hageman et al. method were also conserved (13).
To evaluate the FMB data processing technique, push-pull tests were simulated using
a numerical model. FMB-adjusted concentrations were calculated from the simulated data and
rate constants were derived from FMB-adjusted concentrations.
Errors in derived rate
constants were quantified with respect to reactant and product retardation factors, test
duration, groundwater velocity, first-order rate constant, and dispersivity by comparing
derived rate constants to the rate constants used as inputs in the numerical model. The FMB
technique was then applied to data obtained during an actual field push-pull test designed to
measure the in situ reductive dechlorination rate of trichiorofluoroethene (TCFE) in TCEcontaminated groundwater.
METHODS
Numerical Simulations
Push-pull tests were simulated with the fully implicit, volume-integrated, finite
difference simulator, Subsurface Transport over Multiple Phases (STOMP) (17), with the one-
dimensional linear form of the advection-dispersion equation.
equally spaced grid blocks (0.05 m
x 1 m ><
The mesh consisted of 400
1 m). Bulk density, p, (2.3 kg/L) and porosity, n,
(0.2) values were selected based on measurements conducted on aquifer materials collected at
the field site where the actual push-pull test was conducted (see field push-pull test section).
The groundwater (pore water) velocity, which was selected to represent that measured at the
actual field site, was 0.01 rn/day except where otherwise specified. The groundwater velocity
was established by setting constant pressure boundaries at nodes 1 and 400. The dispersivity,
which was selected to represent that estimated by fitting modeled data to field data from the
site, was 0.1 m except where otherwise specified.
The simulated test solution was injected into node 200 for 125 mm
at 2 L/min to
replicate the injection procedure used during the field push-pull test. Simulated test solutions
contained the sorbing reactant, A, which was allowed to undergo transformation by first-order
kinetics in the aqueous phase only to the product, B. The input value for the first-order rate
constant,
k,
was 0.069 day' except where otherwise specified and was selected because it was
similar to that observed in the field test. The time step used during the injection phase was
0.05 mm. So that the transport behaviors of the sorbing solutes could be compared to that of a
Following the injection, solutes were allowed to undergo transport and transformation for 90
days except where otherwise specified. Time steps were 0.05 days during this phase. To
simulate sampling of the test solution/groundwater mixture, aqueous concentrations at the well
(node 200) were output once per day. Although concentration versus distance data are not
available during an actual field push-pull test, it was output at given times during simulated
push-pull tests to aid in understanding the effects of transport behavior on concentrations at
the single well. An example input file is included in the Supporting Information.
Test set I was designed to determine the accuracy in rates derived using FMB-adjusted
concentrations for three illustrative cases. The three simulations were conducted using a
retardation factor for A,
R(A),
of 5. The retardation factor for B, R(B), was 5, 1.25, and 20
during the three respective tests. Thus,
times greater than
R(B)
R(A)
was equal to R(B) in the first simulation, four
in the second simulation, and four times less than
R(B)
in the third
simulation. Test set II was designed to determine the effects of test duration, groundwater
velocity, dispersivity, and input
k
on the accuracy of derived rates. For each of these
parameters, a series of push-pull tests was simulated in which that single parameter was varied
while the others were held constant. Simulations were conducted with R(A) and
R(B)
set to 5
and 1.25, respectively, and then with R(A) and R(B) set to 5 and 20, respectively. Test set III
was designed to determine the effects of retardation factors on the accuracy of derived rates
for a broader range of R(A) and R(B) combinations than that used in test set I.
Application of FMB to a Field Push-Pull Test
A field push-pull test was conducted in TCE-contaminated groundwater located at the
site of a former chemical manufacturing plant in the San Francisco Bay area where TCE
38
reductive dechlorination has been monitored in recent years (13, /8). The test was conducted
in the C-zone, which is characterized by alluvial fan deposits and is located approximately 6 to
23 m below the ground surface. C-zone groundwater velocities range from 6 to 31 rn/year.
The test solution (.250 L) was injected at a rate of 2 L/min between a pair of inflatable
packers designed to isolate a 1-meter section of the well screen.
Extraction samples were
collected once per week for 84 days. Further details about the field site, experimental design,
and analytical methods are provided elsewhere (13).
A modeling exercise was used to quantitatively evaluate the accuracy of the rate
derived from field data using the FMB technique. A series of simulations was conducted
using the same input parameters used in test sets 1-111. However, the injected test solution
contained TCFE instead of the generic sorbing reactant, A. TCFE was allowed to undergo
sequential transformation to DCFE and CFE (but not to FE since FE was not detected in
extraction samples) by first-order kinetics in the aqueous phase only.
conducted with a range of input
k
values that bracketed the field-derived
k
Simulations were
value. Retardation
factors of 2.05, 1.39, and 1.14 were used for TCFE, DCFE, and CFE, respectively, based on
calculations using
Rl+Kombom
(8)
n
where
K0,
is the organic matter/water partition coefficient of the solute; .f1 and
ph
are the
fraction of organic matter and bulk density of aquifer sediments, respectively; and n is the
porosity of the aquifer. The Estimations Programs Interface Suite (19) was used to assign
K0,
values of 90.5, 33.5, and 12.2 L/kg to TCFE, DCFE, and CFE, respectively. Values used for
f°m
(0.00 1), ph (2.3 kg/L), and n (0.2) were selected based on measurements conducted on
aquifer material collected at the field site. An input k value of 0.0017
day1
was selected to
describe the transformation of DCFE to CFE based on a best fit analysis between modeled and
field data.
RESULTS AND DISCUSSION
Test Set I
Test 1: R(A) = R(B) = 5.
At the end of the simulated test solution injection (time
125 mm), the maximum total concentrations of the co-injected conservative tracer, T. and the
sorbing reactant, A, were at the well (distance = 0 meters) (Figure 3. Ia). The concentration
39
profile of A was narrower than that of T because it sorbed to aquifer sediments during the
injection while T did not sorb. After 90 days of transport and transformation, peaks of A and
its product, B (formed in situ), were centered at an equal distance from the well (Figure 3. Ib)
but were chromatographically separated from the peak of T. This separation occurred because
the transport velocities of A and B were less than that of T. The transport velocity of a solute
equals the groundwater velocity divided by the solute's retardation factor. (Note that the
retardation factor of T is
1.)
The k used as an input in the simulation code for the
transformation of A to B could have been back calculated by integrating total concentrations
of A and B in the aquifer at any time. However, because concentration measurements are
made at the single well location only during an actual field push-pull test, the challenge was to
derive
exclusively from the total concentrations found at the single simulated well (Figure
k
3. 1 c).
FMB-adjusted concentrations
of A and B were
calculated from their total
4
0
0
.
0
a
1.5
0
0
0
0
U
U
a
0
H
0
0.5
H
T
0.0
0
0
-1
1
2
-2
3
-1
Distance (m)
4
0
9
0
0
a
\
0
U
0
U
H
40
60
Time (days)
2
3
A--13
0.8
AFMI3
0.6
S..
80
Aid!
0.4
0.0
0
20
1
1.0
0.2
0
0
0
Distance (m)
BFMB
Bid
t
0
20
40
60
Time (days)
Figure 3.1: Test Set I, Test 1,
(d)
80
R(A) = R(B) = 5. Total (aqueous plus sorbed)
concentrations of the reactant, A, the product, B, and the tracer, T, in the simulated
aquifer at (a) the end of the test solution injection (125 mm) and (b) 90 days. The
vertical line (at distance = 0 days) represents the single well. (c) Total concentrations
and (d) FMB- and ideal transport-adjusted concentrations at the simulated well.
40
a
'.5
a
V
0
1.0
1.0
0.8
A,
0.6
L)
C
0.4
0
A+B
.AFMB
0.5
.E.
H
Bld
02
BFMB
(b)
0.0
0.0
-2
0
-1
1
2
0
3
20
40
60
80
Time (days)
Distance (m)
Figure 3.2: Test Set I, Test 2, R(A) = 5, R(B) = 1.25. (a) Total concentrations of A,
B, and T in the simulated aquifer at 90 days. (b) FMB- and ideal transport-adjusted
concentrations at the simulated well.
concentrations using eq 4 (Figure 3.1 d). Ideally, the FMB technique removes the effects of
transport processes (including sorption) from total concentrations. To test the effectiveness of
the FMB technique, "ideal" transport-adjusted concentrations of A and B were calculated
(Figure 3.1 d).
Ideal concentrations are those that would occur if the reaction were taking
place in the absence of transport processes (e.g. in a batch reactor). Ideal (batch reactor)
concentrations of A and B were calculated at each time point used in the simulation by
[A]t11
where
[A]eai
[A];1
*
deal
(t
\n
(9)
t_1 ) * k / R(A)
is the ideal concentration of A at time, t,,. The value for
equal to the input
k
used in the simulations.
It was divided by
k
used in eq 9 was
R(A)
because when
transformation is only permitted in the aqueous phase, the effective rate constant equals
k
divided by the retardation factor of the reactant. Ideal transport-adjusted concentrations of B
were calculated at each time point as the difference between the initial ideal concentration of
A and the ideal concentration of A at ti,. FMB- and ideal transport-adjusted concentrations of
A and B were identical during this test (Figure 3.ld), indicating that the FMB technique
successfully removed the effects of transport processes from total concentrations.
The performance of the FMB technique was also evaluated by comparing the
k
derived from FMB-adjusted concentrations to the k used as an input in the simulation model.
The best fit slope of ln ([AIItAj,0) versus time (data not shown), where
injected FMB-concentration of A, represents
k/R(A)
[A],0
is the
and was equal to 0.014 day1. Thus, the
derived k (0.069 day1) was obtained by multiplying the slope by
R(A)
and was equal to the
input k (0.069 day1). The error quotient, or the ratio relating the input and derived values for
41
C
10
'.5
1.0
o
C
C-)
0.5
0
0.6
A FMB
04
BFMBr
° 02
H
A+B
AdaJ
0.8
.....
B ideal
-&
(b)
0.0
-2
-1
0
1
2
Distance (m)
3
0
20
40
60
80
Time (days)
Figure 3.3: Test Set I, Test 3, R(A) = 5, R(B) = 20. (a) Total concentrations of A, B,
and T in the simulated aquifer at 90 days. (b) FMB- and idea! transport-adjusted
concentrations at the simulated well.
k,
was 1. The conclusion that the FMB technique can be used to determine accurate in situ
transformation rates when the reactant and product are transported identically was further
supported by results from other simulated tests in which values of R(A) and
R(B)
were equal
(data not shown).
Test 2: R(A,)
5, R(B) = 1.25.
The concentration profiles of I and A at the end of the
test 2 injection phase (data not shown) were identical to those observed at the end of the test 1
injection phase (Figure 3.la). Furthermore, the peak areas of A and B at 90 days (Figure 3.2a)
were identical to those observed at 90 days during test 1 (Figure 3. lb) because the effective
rate constant,
kIR(A),
was identical during the two tests. However, in test 2, peaks for A. B,
and T were chromatographically separated from each other since each solute had a unique
transport velocity. Because B was transported further downgradient during test 2 than during
test 1, its total concentration at the well at 90 days was lower in test 2 than in test 1. As a
result, the ratio of the total concentrations of A to B at the well at 90 days was higher in test 2
than in test 1. Since this ratio is proportional to the FMB-adjusted concentration of A (eqs 45), FMB-adjusted concentrations of A were higher in test 2 (Figure 3.2b) than in test 1 (Figure
3.ld) and deviated from their ideal transport-adjusted concentrations (Figure 3.2b). Likewise,
FMB-adjusted concentrations of B were lower in test 2 than in test 1 and also deviated from
The deviation of FMB-adjusted concentrations of A from ideal
their ideal concentrations.
concentrations caused the
relative to the input
derived
k,
k
k
derived from FMB-adjusted concentrations (0.041 day1) to be low
(0.069 day1). Thus, the error quotient, or the ratio of the input
k
to the
was 1.7. Results from this simulated test indicate that in situ transformation rates
42
estimated from field test data may be low by a factor of less than two when field conditions
are similar to those used in this simulation.
Test 3: R(A) = 5, R(B) = 20.
The concentration profiles of T and A at the end of the
test 3 injection phase (data not shown) were identical to those at the end of the test 1 injection
phase (Figure 3.la). Furthermore, the peak areas of A and B at 90 days (Figure 3.3a) were
identical to those observed at 90 days during test 1 (Figure 3.lb) because the effective rate
constant,
k/R(A),
was identical during the two tests. The peaks of A, B, and T were again
separated from each other chromatographically; however, in this case, the B peak was located
closer to the well than the A peak. The chromatographic separation of A and B peaks during
this test caused the FMB concentrations of A and B to deviate from their ideal transportadjusted concentrations (Figure 3.3b). However, since the B peak was located closer to the
well than the A peak, FMB-adjusted concentrations of A were lower than ideal concentrations
of A and FMB-adjusted concentrations of B were higher than ideal concentrations of B. As a
result, the k derived from FMB-adjusted concentrations of A (0.082 day') was high relative to
the input k (0.069 day1). The error quotient, which in this case was defined as the derived kto
the input k so that it wouldn't have to be expressed as a fraction, was 1.2. The error quotient
obtained in test 3 was smaller than that obtained in test 2 because the degree to which A and B
were chromatographically separated was less in test 3 than in test 2. Results from this
simulated test indicate that in situ transformation rates estimated from field test data may be
high by a factor of less than 1.5 when field conditions are similar to those used in this
simulation.
Test Set II
The effects of test duration, groundwater velocity, input k, and dispersivity on the
accuracy of derived rates were determined for cases in which R(A) and R(B) were 5 and 1.25,
respectively, and 5 and 20, respectively. From this point forward, the error quotient is defined
as the ratio of the input k to the derived k for all cases in which
the ratio of the derived k to the input
k
for all cases in which
R(A)
R(A)
is greater than R(B) and
is less than
R(B).
Error
quotients increased as test duration (Figure 3.4a) or groundwater velocity (Figure 3.4b)
increased since increases in these parameters led to increases in the chromatographic
separation of A and B. Error quotients decreased as the input k increased (Figure 3.4c)
because faster rates allowed more of the conversion of A to B to occur before
chromatographic separation became significant. Error quotients also decreased with
43
2.0
(a)
1.8
10
(b)
R(A)=5,
1.6
)/
C
R(A)=5,
1.4
I-
C
1.2
in
0
20
40
60
80
Test Duration (days)
C
.
C
_-
VR(A)=5,
0.00 0.02 0.04 0.06 0.08 0.10
Groundwater Velocity (rn/day)
4
R(A) =5,
1.8
R(A) =5,
/ R(B)=1.2/
1.25
1.25
(d)
R(A) =5
C
1.6
I-
C
1.2
1.0'
0.0 0.1 0.2 0.3 0.4 0.5
0.6 0.7
0.2
0.0
Input k (day')
0.4
0.6
0.8
Dispersivity (rn)
1.0
Figure 3.4: Test Set II. Error quotients as a function of (a) test duration, (b)
groundwater velocity, (c) input k, (d) dispersivity.
increasing dispersivity (Figure 3.4d) because dispersivity affected the peak widths of A and B.
As A and B peaks became more narrow (i.e. as dispersivity decreased), the overlap of the two
peaks decreased and the ratio of the total concentrations of A to B at the well increasingly
deviated from what it would have been if A and B were transported identically.
Test Set III
The effect of retardation factor values on the accuracy of derived rates was assessed
from test set III simulations. For cases in which
became increasingly larger as the ratio of
R(A)
R(A)
to
was greater than
R(B)
R(B),
error quotients
increased for any given
R(A)
value
(Figure 3.5a). Again, chromatographic separation was responsible for error. For example, the
chromatographic separation of A and B was significantly greater when R(A) and
R(B)
were 50
and 1, respectively, (Figure 3.5b) than when they were 5 and 1.25, respectively (Figure 3.2a).
Error quotients increased as
R(A)
decreased for a given
R(A)
to
R(B)
because differences in transport velocities between A and B increased as
given
R(A)
to
R(B)
ratio.
ratio (Figure 3.5a)
R(A)
decreased for a
44
6
0.8
R(A)=5
R(A)=10
0
0
0.6
0
0
0
C-)
0
02
F-
0.0
0.0
0.5
1.0
1.5
-2
0
-1
logR(A) /RB,)
1
2
3
Distance (m)
1.3
(c)
0
1.5
0
1.2
0
1.0
/
1.1
-.-.-.- R(A)1.25
R(A)=5
0
H
1.0
0.5
0.0
0.0
0.5
1.0
-7
1.5
IogR'B) /R('A)
0
-1
I
2
3
1)istance (rn)
Figure 3.5: Test Set III. (a) Error quotients as they relate to retardation factor when
R(A) > R(B). (b) Total concentrations of A, B, and T in the simulated aquifer at 90
days when R(A) = 50 and R(B) = 1. (c) Error quotients as they relate to retardation
factor when R(A) <R(B). (d) Total concentrations of A, B, and T in the simulated
aquifer at 90 days when R(A) 5 and R(B) = 250.
For cases in which R(A) was less than R(B), error quotients began to level off after an
initial increase as the ratio of R(B) to R(A) increased for any given R(A) value (Figure 3.Sc).
Maximum error quotients were small relative to those observed when R(A) was greater than
R(B)
because the chromatographic separation of A and B was minimal during all cases in
which
R(A)
was less than
R(B).
For example, A and B did not undergo significant
chromatographic separation even when R(A) and
R(B)
were 5 and 250, respectively, (Figure
3.5d). Error quotients increased as R(A) increased for a given R(A) to
R(B)
ratio (Figure 3.5c).
This trend cannot be attributed to differences in transport velocities between A and B since
they increased as R(A) decreased for a given R(A) to R(B) ratio. Thus, R must have played a
different, more important role in affecting errors when R(A) was less than
R(B).
One
explanation is that errors increase with increasing R(A) because R influences the effective rate
constant
set II).
(k/R)
and the effective dispersivity (dispersivity/R), which both influence error (test
45
FIELD APPLICATION
Concentrations of TCFE and its transformation products, DCFE (cis- and trans-),
CFE, and FE were measured in extraction samples collected during a field push-pull test. The
test was conducted by injecting a test solution containing TCFE into TCE-contaminated
groundwater. Total (aqueous plus sorbed) concentrations for each solute except FE, which
was not detected, were calculated from their measured concentrations using equations
analogous to eqs 6 and 7. The values used in these calculations for K(,fl, were given in the
methods section and the values used
forf0,
(0.00 1), n (0.2), and P (2.9 kgIL) were determined
from aquifer material collected at the field site. FMB-adjusted concentrations were calculated
for each solute using eq 4 (Figure 3.6a). The best fit slope of in ([TCFE}FMB/[TCFE]J,O)
versus time (Figure 3.6b), where [TCFE]F/0 is the injected FMB-adjusted concentration of
TCFE, was 0.079 day' between 0 and 30 days. The field-derived k was calculated by
multiplying this slope by the retardation factor of TCFE, R(TCFE), which was 2.05 (see
methods section). Hence, the field-derived k for the transformation of TCFE to DCFE was
0.16 day'.
Error estimates from the sensitivity analysis performed in test sets 1-111 were then used
to qualitatively assess the accuracy of the field-derived k. First, it is likely that the fieldderived k is underestimated relative to the actual in situ k since the retardation factor of TCFE
is greater than that of its products. In addition, the magnitude of the error is likely to be less
than a factor of two since data from the first 30 days of the test were used to estimate k(Figure
Of(b
2
A
A
cts-DCFE
trans-DCFE
CFE
O2O4O8O
Time (days)
-2
Field Data
Regression Line
Time (days)
Figure 3.6: (a) FMB-adjusted concentrations of TCFE and its products, indicating
that reductive dechlorination of TCFE occurred during this field push-pull test
conducted in TCE-contaminated groundwater and (b) the plot from which the derived
first-order rate constant was determined.
46
3.4a), the groundwater velocity was -0.01 mlday (Figure 3.4b); the estimated input (or in situ)
k was
0.16
day1
(Figure 3.4c), and the dispersivity in the aquifer was -..0.l m (Figure 3.4d).
The field-derived k is also expected to be in error by a factor of less than two than since the
log R(TCFE)/R(DCFE) value was 0.17 (Figure 3.5a).
The accuracy of the field-derived k was assessed quantitatively by simulating pushpull tests with a test solution containing TCFE. Simulations were conducted with a range of
input k values (0.07-0.26 day1) that bracketed the field-derived k. Derived k values were
determined for each simulation using the FMB technique. The empirical relationship between
the model input and derived k values for the transformation of TCFE to DCFE was
input k
1. 1 * derived k + 0.0080,
(10)
which indicates that the field derived k was underestimated relative to the actual in situ k by a
factor of 1. 1, or 10%. If this error quotient is applied to the field-derived k, the estimated in
situ k becomes 0.18
day1.
The potential effects of non-equilibrium sorption, aquifer property heterogeneities,
and reaction rate heterogeneity on derived rates were not evaluated in this study. These issues
will be investigated in future research projects. Nonetheless, to the best of our knowledge, the
FMB technique makes it possible to estimate in situ transformation rates of sorbing solutes
from push-pull test data for the first time. Moreover, errors in rates derived using the FMB
technique can be quantified by coupling the technique to computer modeling.
ACKNOWLEDGEMENTS
We thank Mark White from the Pacific Northwest National Laboratory for writing the
STOMP code used in this project. We are also grateful to Brian Wood, Roy Haggerty, Brian
Davis, and Melora Park for their contributions. This project was funded by grant number
lP42 ES 10338 from the National Institute of Environmental Health Sciences (NIEHS). with
funds from the U.S. Environmental Protection Agency.
LITERATURE CITED
(1)
McAllister, P. M.; Chiang, C. Y. Ground Water Monit. Rev. 1994, 14, 161-173.
(2)
Weaver, J. W.; Wilson, J. T.; Kampbell, D. H. Proceedings of the Symposium on
Natural Attenuation of Chlorinated Organ/cs in Ground Water; Office of Research
and Development, Ed.; U.S. EPA: Dallas, Texas, 1997; pp 71-75.
(3)
Wiedemeier, T. H.; Swanson, M. A.; Wilson, J. T.; Kampbell, D. H.; Miller, R. N.:
Hansen, J. E. Ground WaterMonil. Rev. 1996, 16, 186-194.
47
(4)
US EPA Office of Research and Development "Technical Protocol for Evaluating
Natural Attenuation of Chlorinated Solvents in Ground Water," 1998 EPA/600/R98/128.
(5)
Buscheck, T. E.; Alcantar, C. M. In Intrinsic Bioremediation; Robert E. Hinchee,
Wilson, John T., Downey, Douglas C., Eds.; Battelle Press: Columbus, 1995; pp 109116.
(6)
McNab, W. W. J.; Dooher, B. P. Ground Water 1998, 36, 983-987.
(7)
Washington, J. W.; Cameron, B. A. Environ. Toxicol. Chem. 2001, 20, 1909-1915.
(8)
Istok, J. D.; Humphrey, M. D.; Schroth, M. H.; Hyman, M. R.; O'Reilly, K. T. Ground
Water 1997, 35, 619-63 1.
(9)
Schroth, M. H.; Istok, J. D.; Conner, G. T.; Hyman, M. R.; Haggerty, R.; O'Reilly, K.
T. Ground Water 1998, 36, 924-937.
(10)
Pombo, S. A.; Pelz, 0.; Schroth, M. H.; Zeyer, J. FEMS Microbiol. Ecol. 2002, 41,
(11)
259-267.
Kleikemper, J.; Schroth, M. H.; Sigler, W. V.; Schmucki, M.; Bernasconi, S. M.;
Zeyer, J. App!. Environ. Microbiol. 2002, 68, 1516-1523.
(12)
Schroth, M. H.; Kleikemper, J.; Bollinger, C.; Bernasconi, S. M.; Zeyer, J. I Contain.
Hydrol. 2001, 51, 179-195.
(13)
Hageman, K. J.; Istok, J. D.; Field, J. A.; Buscheck, T. E.; Semprini, L. Environ. Sci.
Technol. 2001, 35, 1729-1735.
(14)
Haggerty, R.; Schroth, M. H.; Istok, J. D. Ground Water 1998, 36, 3 14-324.
(15)
Snodgrass, M. F.; Kitanidis, P. K. Ground Water 1998, 36, 645-650.
(16)
Vancheeswaran, S.; Hyman, M. R.; Semprini, L. Environ. Sci. Technol. 1999, 33,
(17)
White, M. D.; Oostrom, M. "Stomp: Subsurface Transport over Multiple Phases.
(18)
Buscheck, T. E.; O'Reilly, K. T.; Hickman, G. In In Situ and on-Site Bioremediation;
Battelle Press: Columbus, OH, 1997; Vol. 3, pp 149-154.
(19)
Syracuse Research Corporation "Estimations Programs Interface Suite," 2000,
2040-2045.
User's Guide," Pacific Northwest National Laboratory, 1997 PNNL-12034.
www.epa.gov/oppt/exposure/docs/episuite.htm.
SUPPORTING DFORMATION: EXAMPLE INPUT FILE FOR USE WITH STOMP
# ------------------------------------------
-Simulation Title Card
--------
1,
Forced Mass Balance Test 080802j,
Kimberly Hageman,
48
Oregon State University,
08 August 2002,
12:30 pm PDT,
6,
Simulate T(racer), A, and B transport with I st-order-aq,
P diff= 9.8 Pa, HC = 40 rn/day, pw velocity = 0.01 rn/day,
Dispersivity = 0.1 m, 0.004 m,
R(A) = 5, R(B) = 5,
Storativity = 0,
Half-life for A = 10 days,
Solution Control Card
# ------------------------------------------
Normal,
Water w/TVD Transport,
2,
0.0,min, 1 25,min,0.05,min,0.05,min, 1.0,8,1 .e-06,
1 25,min,90,day,0.05,rnin,0.05,day, 1.25,8,1 .e-06,
10000,
0,
'-Grid Card
# -------------------Cartesian,
400,1,1,
0,m,400@0.05,rn,
0,m,1,m,
0,m,1,rn,
# -------------------------------
-Rock/Soil Zonation Card
# ------------------------------1,
RichSoil,1,400,1,1,1,1,
Mechanical Properties Card
-------
RichSoil,2900,kg/m"3 ,0 .2,0.2,0,,Millington and Quirk,
# ---------------------------------
-Hydraulic Properties Card
RichSoil,40,hc rn/day,,,,,
ft
49
-Saturation Function Card
RichSoi!,Nonhysteretic van Genuchten,0 .025,1 /cm,3 .0,0.05,,
# ------------------------------------------
-Aqueous Relative Permeability Card
RichSoil,Mualem,,
--------
# ------------------------------------
-Solute/Fluid Interaction Card
3,
T,Conventional, 1 .0E9,mA2/s,Continuous, 1 .OE 1 0,yr,
A,Conventional, 1 .OE-9,m"2/s,Continuous, 1 0,day,
B,Conventional, 1 .0E-9,rn''2/s,Continuous, I .OE 1 0,yr,
I,
A,B, 1,
--Solute/Porous Media Interaction Card
# -----------------------------------------RichSoil,0. 1 ,m,0.004,m,
T,0, liter/kg,
A,0.3448,Iiter/kg,
B4O.3448,liter/kg,
Initial Conditions Card
Gas Pressure,Aqueous Pressure,
4,
Aqueous Pressure,25 8185 .8,Pa,-0.49, 1/rn,,,,, 1,400,1,1,1,1,
Solute Aqueous Conc.,T,0, I /liter,,,,,,, 1,400,1,1,1,1,
Solute Aqueous Conc.,A,0, 1 /liter,,,,,,, 1,400,1,1,1,1,
Solute Aqueous
Conc.,B4O, 1 /liter,,,,,,, 1,400,1,1,1,1,
--Boundary Conditions Card
2,
West,Dirichlet Aqueous,Outflow,Outflow,Outflow,
1,1,1,1,1,1,1,
0.,s,258 185 .8,Pa,0.,,,,,,,
East,Dirichlet Aqueous,Outflow,Outflow,Outflow,
400,400,1,1,1,1,1,
0.,s,258 I 76.0,Pa,,,,,,,
50
Source Card
3,
Aqueous Vo!umetric,200,200, 1,1,1,1,2,
0.,min,2.0,liter/min,
1 25,min,2.0,liter/min,
Solute,T,200,200, 1, 1, 1,1,2,
0.,min,30,1/min,
1 25,min,3 0,1 1mm,
Solute,A,200,200, 1, 1, 1, 1,2,
0.,min,30, 1/mm,
1 25,min,30, 1/mm,
# -------------------------
-Output Control Card
3,
100,1,1,
200,1,1,
3 00, 1, 1,
10,1 0,day,m,,7,7,7,
3,
Solute Aqueous Conc.,T, 1/liter,
Solute Aqueous Conc.,A,1/liter,
Solute Aqueous Conc.,B,1/liter,
4,
I 25,min,
3 0,day,
60,day,
90,day,
3,
Solute Aqueous Conc.,T,l/liter,
Solute Aqueous Conc.,A, 1/liter,
Solute Aqueous Conc.,B,1/Iiter,
51
CHAPTER 4. QUANTIFYING THE EFFECTS OF CHEMICAL AMENDMENTS ON
IN SITU REDUCTIVE DECHLORINATION RATES
Kimberly I Hageman a Jennfer A. Field', Jonathan D. IstOkc, and Lewis Semprinf
aDepament of Chemistry, bDepament of Environmental and Molecular Toxicology,
Department of Civil, Construction, and Environmental Engineering
Oregon State University, Corvallis, OR 97331-7301
52
ABSTRACT
In situ methods are needed to evaluate the effectiveness of chemical amendments at
enhancing reductive dechlorination rates in groundwater that is contaminated with the priority
pollutant, trichloroethene (TCE). In this communication, a method that utilizes single-well,
"push-pull" tests to quantify the effects of chemical amendments on in situ reductive
dechlorination rates is presented and demonstrated. Five push-pull tests were conducted in
each of five monitoring wells located in a TCE-contaminated aquifer at the site of a former
pesticide-manufacturing site. Rates for the reductive dechlorination of the fluorinated TCEsurrogate, trichlorofluoroethene (TCFE), were measured before (test 1) and after (test 5) three
successive additions (tests 2-4) of fumarate, which was selected to stimulate the growth and
activity of indigenous microorganisms with the metabolic capability to reduce TCFE and
TCE.
In three wells, first-order rate constants for the reductive dechlorination of TCFE
increased by 8.9 to 120 times following fumarate additions.
In two wells, reductive
dechlorination of TCFE was observed after fumarate additions but not before. The
transformation behavior of fumarate was also monitored following each fumarate addition.
Correlations between the reductive dechlorination of TCFE and the reduction of fumarate to
succinate were observed, indicated that these reactions were supported by similar
biogeochemical conditions at this site.
53
INTRODUCTION
Significant research efforts have been devoted to the development of in situ
bioremediation as an approach for remediating groundwater contaminated with the priority
pollutant, trichloroethene (TCE) and other chlorinated aliphatic hydrocarbons. In anaerobic
environments, this approach depends on the metabolic capability of indigenous subsurface
microorganisms to catalyze the reductive dechlorination of TCE to the dichloroethene (DCE)
isomers, chioroethene (CE), and ethene (1-3) (Figure 4.la). Engineered approaches are
needed where natural attenuation does not result in the complete conversion of TCE to ethene
or where rates are too slow to meet risk management goals.
A common approach for enhancing in situ reductive dechlorination is to stimulate the
growth of indigenous dechlorinating microorganisms with the addition of chemical
amendments. A wide variety of chemicals and chemical mixtures have been evaluated for
their suitability as amendments for enhancing reductive dechlorination. Lee et al. reviewed
results from laboratory tests that were designed to assess the effectiveness of potential
amendments such as complex organic mixtures (molasses, wastewater, cheese whey permeate,
corn steep liquor, manure tea), metabolic intermediates (benzoate, lactate, propionate, acetate,
butyrate), alcohols (methanol, ethanol), molecular hydrogen, sulfate, nitrate, vitamins, and
micronutrients (4, 5). While many of these amendments were effective, disadvantages were
associated with each and none were universally effective at stimulating reductive
dechlorination in groundwater from all sites. To the best of our knowledge, fumarate (trans-
(a)
CL
,Cl
Cl,
(b)
CL
TCE
c:H
Irans
-DCE
ci.s-DCE
/
1,1-DCE
TCFE
H
Cl'
trans
Cl
'F
-DCFE
l,l-CFE
H
:
C=C
,Cl
'F
CL
Cl
H
'F
cis-DCFE
Un
(Z)-CFE
Cl
Cl'
H
'F
l,1-DCFE
£L-CFE
Ethene
Figure 4.1. Analogous reductive dechlorination pathways for (a) TCE and (b) its
fluorinated surrogate, TCFE. Predominant isomers and pathways are indicated by
underlines and heavy arrows.
54
0
0
II
0
H
H
Fumarate
+2e+ 2H
0
0
H
H
I
I
I
I
0
,ccccs..
0
II
H
II
H
Succinate
Figure 4.2. Reduction of fumarate to succinate.
1,2-ethenedicarboxylate) has not previously been tested as a chemical amendment for
enhancing reductive dechlorination rates. There is evidence that a number of dechiorinating
microorganisms use fumarate as an alternative electron acceptor (6-13) and that certain
dechlorinating organisms grow faster on fumarate than on chlorinated ethenes (11). Some
dechlorinating microorganisms are also known to use fumarate (14) or succinate (1,2ethanedicarboxylate) (9), which is produced from the reduction of fumarate (Figure 2), as
electron donors during reductive dechlorination. Hence, one objective of this work was to
evaluate the effectiveness of fumarate at enhancing in situ reductive dechlorination rates.
The effectiveness of a chemical amendment is generally evaluated by comparing
reductive dechlorination rates measured with and without the chemical amendment
laboratory experiments with pure or mixed cultures of microorganisms.
in
Experiments
conducted in actual aquifers are not as common because field experiments, especially those
involving well-to-well tests, are perceived as complicated, expensive, and/or time-consuming
in comparison to laboratory experiments. In addition, in situ transformation rates are difficult
to measure because solute concentrations in groundwater are affected by both transformation
and transport (advection, dispersion, and sorption) processes. Nevertheless, the need for field
methods that can be used to evaluate the effectiveness of chemical amendments becomes
increasingly apparent as concerns about discrepancies between laboratory and field results
mount (15-1 7). Field pilot tests designed to determine in situ reductive dechlorination rates in
the presence of chemical amendments (e.g. acetate, nitrate, and sulfate) were reviewed by Lee
et al. (5). In all but one of the thirteen reported pilot tests, in situ reductive dechlorination
rates were determined using multi-well tests that involved the recirculation of amended
groundwater between injection and extraction wells.
Hageman et al. recently described an alternative method for determining in-situ
reductive dechlorination rates (18) that utilizes single-well, "push-pull" test technology (19).
Push-pull tests were conducted in a TCE-contaminated aquifer at the site of former pesticide-
55
manufacturing plant in the San Francisco Bay area. Each push-pull test was conducted by
injecting ("pushing") an aqueous test solution into the saturated zone via a monitoring well.
Test solutions contained bromide, which served as a conservative tracer; trichiorofluoroethene
(TCFE), which served as a surrogate for TCE; and in some cases, formate, which is a common
chemical amendment used as an electron donor. TCFE was selected as a surrogate for TCE
based on evidence that it undergoes reductive dechlorination by a pathway analogous to that
of TCE while retaining the fluorine label (Figure 4.lb) (20). TCE itself was not injected
because mixing of the injected test solution with native groundwater would have rendered it
impossible to distinguish injected and background TCE.
Following the injection phase,
samples of the test solution/groundwater mixture were periodically extracted ("pulled") from
the single well and analyzed for TCFE, transformation product, and tracer concentrations.
TCFE reductive dechlorination rates were determined from measured concentrations that had
been adjusted with a data processing technique designed to remove the effects of transport
processes. Hageman et al. discussed reductive dechlorination rates obtained in three wells at
the site (18). In one of those wells, the effect of formate on in situ reductive dechlorination
rates was quantified by comparing the rate obtained during a test conducted by co-injecting
TCFE and formate to the rate obtained in a later test conducted by injecting TCFE without
formate.
There are a number of advantages associated with using push-pull tests instead of
multi-well tests to evaluate chemical amendments. For instance, push-pull tests take less time
to conduct than multi-well tests since injected solutes do not have to be transported between
wells. Because
push-pull
tests are time-efficient,
push-pull
tests can be repeated in a single
well with and without a co-injected chemical amendment or before and after a suite of
chemical amendment additions to the well. At sites where a number of monitoring wells exist,
push-pull tests can be conducted simultaneously in different wells to determine the effects of
different amendments on transformation rates or to assess spatial variability in transformation
rates. Push-pull tests are cost-effective relative to multi-well tests because fewer groundwater
wells, which are expensive to construct, are needed. Moreover, a revised form of the data
processing technique designed to remove the effects of transport processes from measured
concentrations of sorbing solutes was recently presented (21). The availability of this data
processing technique makes it possible to obtain in situ transformation rates from push-pull
test data for a larger selection of reactants.
56
The objective of this study was to quantify the effects of a specific chemical
amendment, namely fumarate, on in situ reductive dechlorination rates of TCFE using push-
pull tests. To this end, TCFE reductive dechlorination rates were measured before and after
three consecutive additions of fumarate in five wells.
Additionally, the transformation
behavior of fumarate was monitored after each fumarate additions so that potential
correlations between reductive dechlorination and fumarate transformation behavior could be
assessed.
EXPERIMENTAL SECTION
Chemicals
Trichiorofluoroethene (TCFE)(97% pure, containing 0.1% cis-DCFE and 0.3% transDCFE),
cis/trans- 1
,2-dichloroethene (DCFE) (98%), and E/Z- 1 -chloro-2-fluoroethene (97%)
were obtained from SynQuest Laboratories, Inc. (Alachua, FL). Fluoroethene (FE) (98%) was
obtained from Lancaster Synthesis (Peiham, NH).
Sodium fumarate (98%) and sodium
succinate (99%) were obtained from Aldrich Chemical Co., Inc. (Milwaukee, WI). Potassium
bromide (99.7%) and sodium formate (99.6%) were obtained from Fisher Scientific (Fair
Lawn, NJ). For use as an internal standard, 1-chioropropane was obtained from Matheson
Company (Cincinnati, OH).
Site Description
Push-pull tests were conducted at the site of a former pesticide-manufacturing plant in
the San Francisco Bay area where reductive dechlorination has been monitored in recent
years (18, 22, 23). Tests were conducted in two distinct aquifer zones. The A-zone is an
unconfined shallow layer composed mainly of placed fill over Bay Mud. The water table lies
within 3 meters of the ground surface. The groundwater velocity ranges from 1.5 to 6 meters
per year. The C-zone underlies the Bay Mud and is characterized by alluvial fan deposits
located approximately 6 to 23 meters below the ground surface. Groundwater velocities range
from 6 to 3 1 meters per year.
Push-Pull Tests
A series of five push-pull tests were conducted in each of two A-zone wells (1 OA and
9A) and in each of three C-zone wells (1SC, 16C, and 21C) between December 1999 and
February 2001. The objective of test 1 was to measure initial rates for TCFE reductive
dechlorination. Therefore, injected test solutions contained TCFE and bromide (Table 4.1),
which served as a conservative tracer. Fumarate was included in the test solution injected into
57
well 9A but not 1OA so that reductive dechlorination rates could be compared with and
without co-injected fumarate. Formate was included as an electron donor in test solutions
injected into wells 15C and 16C because results from a previously conducted screening
study (23) indicated that reductive dechlorination was limited by electron donor availability in
the C-zone. The objectives of tests 2-4 were to amend the TCE-contaminated groundwater
with fumarate and then to monitor the fumarate transformation behavior following each
addition. Thus, test solutions for tests 2-4 contained fumarate and bromide (Table 4.1). The
objective of test 5 was to measure reductive dechlorination rates after the fumarate additions
so that rate changes due to fumarate could be quantified.
Therefore, TCFE, bromide,
fumarate, and formate were injected in test 5 at concentrations similar to those used in test 1
(Table 4.1).
The experimental design was identical to that described previously (18). Briefly, test
solutions were prepared on-site by adding bromide and fumarate or formate (Table 4.1) to
argon-sparged tap water. In tests I and 5, a concentrated aqueous solution of TCFE was
metered into the injection line. The TCFE solution was stored and pumped from a collapsible
metallized-film gas-sampling bag to prevent volatilization loss during injection. Following
the injection, samples of the test solutionlgroundwater mixture were collected approximately
once per week for up to 85 days during tests I and 5 and on a varying schedule, with a
maximum of 4 samples per day, for up to 20 days during tests 2-4.
Analytical Methods
Chromatographic separation and detection of TCFE, DCFE, CFE, and FE were
achieved using a gas chromatography/mass spectrometry method that was previously
described (18).
The one difference was that purge and trap was used as the analyte
introduction method with test 5 samples. The purge and trap system was composed of a
Tekmar-Dohrmann (Cincinnati, OH) 3100 sample concentrator and an AQUA Tek 70 liquid
analyzer.
Chromatographic separation and detection of organic acids in test series 1-4
samples were performed on a Waters Alliance (Milford, MA) high performance liquid
chromatograph equipped with a photodiode array detector. Separations were achieved on a
Phenomenex (Torrance, CA) Luna C 18 column. Test series 5 samples were analyzed on a
Dionex (Sunnyvale, CA) DX-320 ion chromatograph equipped with a conductivity detector
and AS 11 column. Bromide concentrations were measured with either a Dionex DX- 120 or
00
Table 4.1. Test Solution Compositions
Well
1OA
Test 1
(12/99)
Test 2
(9/00)
(10/00)
Test 3
Test 4
(12/00)
Test 5
(2/01)
13 tM TCFE
0.81 mM fumarate
1.3 mM bromide
0.12 mM fumarate
1.3 mM bromide
0.75 mM fumarate
1.4 mM bromide
1.1 mM bromide
1.3 mM bromide
9.4 tM TCFE
9A
5.3 pM TCFE
1.3 mM bromide
0.75 mM fumar ate
0.87 mM fumarate
1.3 mM bromide
0.13 mM fumarate
1.3 mM bromide
0.74 mM fumarate
0.65 mM bromide
15C
16 P.MTCFE
1.3 mM bromide
8.3 mM formate
0.79 mM fumarate
1.2 mM bromide
0.13 mM fumarate
1.3 mM brom ide
0.74 mM fumarate
1.4 mM bromide
15 JIM TCFE
1.3 mM bromide
6.2 mM formate
16C
14 JJ.MTCFE
1.3 mM bromide
8 mM formate
0.81 mM fumarate
1.2 mM bromide
0.14 mM fumarate
1.4 mM bromide
0.73 mM fumarate
1.3 mM bromide
11 JJ.MTCFE
1.3 mM bromide
6.3 mM formate
21C
33 JIM TCFE
1.4 mM bromide
0.81 mM fumarate
1.3 mM bromide
0.12 mM fumarate
1.3 mM bromide
0.70 mM fumarate
0.98 mM bromide
29 JIM TCFE
1.3 mM bromide
4.3 .tM TCFE
1.3 mM bromide
0.70 mM fumarate
59
Dionex DX-320 ion chromatograph equipped with a conductivity detector and Dionex ASI4
or AS 11 column.
Data Analysis
In situ rates for the reductive dechlorination of TCFE were determined by removing
the effects of transport processes from measured aqueous concentrations of TCFE using a data
processing technique called "forced mass balance" (FMB) (21).
TCFE transformation
products were also treated with the FMB technique so that the distribution of products formed
in situ could be readily compared between tests. The FMB technique was selected over other
available data processing techniques (24, 25) because it was designed for use with sorbing
solutes. Short-term transport tests conducted at the selected site (18) as well as calculated
retardation factor values for TCFE indicated that sorption could affect TCFE transport in both
A- and C-zones. The retardation factor (R) values calculated for TCFE in the A- and C-zones
were 11.5 and 2.05, respectively. These values were calculated (26) using a
K011, value
for
TCFE of 90.5 L/Kg, bulk density of 2.32 KgIL, aquifer porosity of 0.2, and fraction organic
matter values of 0.01 (A-zone) and 0.00 1 (C-zone). The K, value was estimated using the
Estimations Program Interface Suite (27) while the other values were selected based on
measurements made on aquifer solids collected at the site.
Briefly, the FMB technique is conducted by first calculating the mass of the solute in
the aqueous sample obtained during the extraction-phase of the push-pull test. The second
step is to calculate the mass of solute in the sorbed phase associated with the extraction
sample. The third step is to calculate the total (aqueous plus sorbed) concentration of the
solute by dividing the sum of the aqueous-phase and sorbed-phase masses by the sum of the
volumes of the aqueous extraction sample and the associated sorbed-phase. Finally, FMBadjusted concentrations are obtained by dividing each total concentration by the adjustment
factor,
where
and
are the sums of the total concentrations of TCFE and its
transformation products in the extraction sample and in the injected test solution, respectively.
TCFE transformation rates are then determined from its FMB-adjusted concentrations. The
validity of the FMB technique was evaluated by quantifying errors in rates derived by
applying FMB to push-pull test data generated by a numerical model (21). Additionally,
numerical modeling was used to quantify the error in the transformation rate obtained from
push-pull test data obtained during test 5 conducted in well 15C from this study. Since the
error analysis indicated that the in situ rate for the reductive dechlorination of TCFE was
60
underestimated relative to the true in situ rate by 10%, all rates reported herein are expected to
be underestimated by a similar magnitude.
The effects of transport processes were removed from measured aqueous
concentrations of fumarate, succinate, and formate using a data processing technique for use
with nonsorbing solutes hereafter referred to as "tracer-normalization" (24, 25). Sorption was
assumed to have a minimal effect on fumarate, succinate, and formate concentrations since
these solutes are negatively charged and highly water-soluble.
Tracer-normalized
concentrations for each solute were obtained by dividing their measured aqueous
concentrations by the adjustment factor,
[Br]/[Br]0, where [Br] and
[Br]0
are the measured
bromide concentrations in an extraction sample and in the injected test solution, respectively.
RESULTS AND DISCUSSION
Reductive Dechlorination Rates and Product Distribution Ratios
Reductive dechlorination of TCFE occurred following its first injection into well I OA
(test 1, Table 4.1) as indicated by decreasing aqueous TCFE concentrations and increasing
aqueous cis-DCFE concentrations (Figure 4.3a). Trans-DCFE and CFE were also detected at
relatively low concentrations (< 0.05 tM). The FMB technique (see methods section) was
used to obtain adjusted concentrations of TCFE and its transformation products that reflect the
effects of transformation and not transport (Figure 4.3b). The first-order rate constant for the
transformation of TCFE to DCFE was determined by plotting In ([TCFE}F/[TCFE]BO)
values versus time (Figure 4.3c), where [TCFE]
and
[TCFE],0
are the FMB-adjusted
concentrations of TCFE in an extraction sample and in the injected test solution, respectively.
If it is assumed that reductive dechlorination occurs in the aqueous phase only (and not in the
sorbed phase), then the best-fit slope of this plot represents the first-order rate constant (k)
divided by the retardation factor (R) of TCFE. Hence, the slope (0.00 10 day') was multiplied
by the R of TCFE (11.5, see methods section) to obtain an in situ reductive dechlorination rate
of 0.012
day1
(Table 4.2). FMB-adjusted concentrations were also plotted in a stacked area
graph (Figure 4.4a) so that the distribution ratios of TCFE and its products could be more
easily visualized.
(a)
15
O.O5
U,
20
o:o
!
0
20
0
40
60
Time (Days)
80
0.00
1:
0
20
40
60
Time (Days)
80
(c)
TCFE
cis-DCFE
-0.02
L)
S
(b)
U
10
,-
.-. -.
-30
61
-0.04
A
o
frans-DCFE1
CFE
kIR
:
Regression Line
-OH)
0
20
40
60
80
Time (Days)
1 in Well bA.
(a) Aqueous measured concentrations illustrating
concentration changes due to transformation and transport processes, (b) forced mass balance
(FMB)-adjusted concentrations (see data analysis section) illustrating concentration changes
due to transformation only and (c) the plot used to determine the first-order rate constant (k).
Figure 4.3. Test
After three successive additions of fumarate (tests 2-4), which were made 1-2 months
apart from each other, TCFE was injected into well 1OA for a second time (test 5). Reductive
dechlorination of TCFE to cis-DCFE, trans-DCFE, CFE, and FE occurred (Figure 4.4b) with
TCFE being transformed at a maximum rate of 1.4 day' between days 30 and 71 (Table 4.2).
Thus, the maximum transformation rate was 120 times greater after fumarate additions than
before. Moreover, the extent of dechlorination increased, as indicated by the detection of the
less chlorinated products. The formation of FE had important implications for bioremediation
at this site because it is the fluorinated analog of ethene (18, 28) (Figure 4.1) and ethene is the
desired end product for reductive dechlorination due to its lack of toxicity.
62
Table 4.2. TCFE Transformation Rates
Maximum First-Order Rate Constants for
TCFE Transformation (day)a
Test 1
(12/99):
Test 2
(2/01):
1OA
0.012
(0-80 days)
1.4
(30-7 1 days)
9A
0.022
(0-84 days)
(7-23 days)
15C
0.018
(5 5-82 days)
0.16
(0-3 0 days)
16C
--
0.047
(49-84 days)
21C
--
0.14
(16-49 days)
Well
1.8
aTransformation rates were calculated from FMBadjusted concentrations (see data analysis section).
Following the first injection of TCFE into we!! 9A (test 1), TCFE underwent reductive
dechlorination primarily to cis-DCFE although irans-DCFE was also formed (Figure 4.4c).
The rate of TCFE transformation was 0.022 day' (Table 4.2). Note that fumarate was coinjected with TCFE during the initial test conducted in well 9A but not during the initial test
conducted in well I OA. However, the co-injection of fumarate did not appear to affect initial
reductive dechlorination rates since initial rates in wells 9A and 1OA were similar (Table 4.2).
After three additional injections of fumarate (tests 2-4) to well 9A, TCFE and fumarate were
co-injected into well 9A for a second time (test 5). TCFE underwent reductive dechlorination
primarily to cis-DCFE although trans-DCFE and CFE were also formed (Figure 4.4d). TCFE
was transformed at a maximum rate of 1.8 day' between days 7 and 23 (Table 4.2). Thus, the
maximum transformation rate was 82 times greater during test 5 than during test 1. While
TCFE transformation rates obtained during test 5 in wells I OA and 9A were similar,
the extent of dechlorination observed in these wells was remarkably different. In well 9A, cis-
DCFE was formed almost exclusively while in well 1 OA, the dechlorination of TCFE to FE
63
-' 35
30
.,
-
25
15
20
a)
20
10
15
a)
10
5
C.)
(a) 1OA: Test 1
C-)
0
20
(b) 1OA:Test5
U
(1
40
60
Time (Days)
80
0
20
40
60
Time (Days)
80
10
10
,-
a)
o
5
C.)
0
20
4
°
2
C.)
(c) 9A: Test 1
U
a)
40
60
Time (Days)
TCFE
" cis-DCFE
80
(d) 9A: Test 5
0-
0
trans-DCFE
\ \ \ CFE
20
40
60
Time (Days)
80
FE
Figure 4.4.
Stacked area plots of FMB-concentrations depicting reductive
dechlorination of TCFE following the first (test 1) and second (test 5) injections of
TCFE in A-zone wells.
was observed. Possible explanations are that (a) the co-injection of fumarate with TCFE
inhibited further dechlorination or that (b) the microbial populations initially available for
stimulation by fumarate were different in the two wells even though the wells were located
within 25 m of each other and in the same horizontal subsurface zone.
Reductive dechlorination was not observed until 55 days after the first injection of
TCFE into well 15C (test 1) (Figure 4.5a). At that point, TCFE underwent reductive
dechlorination primarily to cis-DCFE although trans-DCFE was also formed. The maximum
rate of TCFE transformation was 0.018 day' (Table 4.2). To increase the likelihood that
reductive dechlorination would occur, formate was co-injected as an electron donor with
TCFE during this test. Co-injected formate was completely degraded by day 27 (data not
64
8
4
o2
a)
a)
U
U
0
20
(b) 15C: Test 5
U
(a) 15C: Test 1
40
60
Time (Days)
I
80
20
40
60
Time (Days)
80
6
&
.-&
3
a)
U!
U
[(c) !6C: Test 1
(1
I
(d) 16C: Test 5
(1
0
20
40
60
Time (Days)
80
0
20
40
60
Time (Days)
80
12
14
10
6
6
U
4
2
(e)21C: Test!
U
0
20
a)
4
U
2
(f)2!C: Test 5
()
40
60
Time (Days)
TCFE
80
'/7' cis-DCFE
0
20
40
60
Time (Days)
80
lrans-DCFE
Figure 4.5.
Stacked area plots of FMB-concentrations depicting reductive
dechlorination of TCFE following the first (test 1) and second (test 5) injections of
TCFE in C-zone wells.
shown), indicating that an active community of formate-utilizing organisms was present.
However, acetate, which would have been fonned during acetogenesis of injected formate,
was not detected. After three successive injections of fumarate (tests 2-4), TCFE and formate
65
were co-injected into well I SC for a second time (test 5). Reductive dechlorination of TCFE
to primarily cis- and trans-DCFE occurred (Figure 4.5b) although CFE was also formed.
TCFE was transformed at a maximum rate of 0.16 day' between days 0 and 30 (Table 4.2).
Thus, the maximum transformation rate was 8.9 times greater after fumarate additions than
before. However, it may be of greater significance that the transformation reaction began
immediately during test 5 instead of being delayed until late in the test as it was during test 1.
Co-injected formate was not detected in any samples, indicating that the formate-utilization
rate also increased between tests. This observation is consistent with the expectation that
electron-donor utilization and reductive dechlorination rates should increase together. Again,
acetate was not detected.
In contrast to what was observed in all previously-described wells, reductive
dechlorination did not occur after the first injection of TCFE into well I 6C (test 1) (Figure
4.5c). Formate, which was again co-injected as an electron donor with TCFE, underwent
degradation slowly such that it was still detected in samples collected at the end of the test
(data not shown). This observation is consistent with the expectation that a slow formateutilization rate would accompany a slow or non-detectable TCFE reductive dechlorination
rate. Acetate formation via the potential acetogenesis pathway was not detected. After three
successive additions of fumarate (tests 2-4), TCFE and formate were co-injected into well 16C
for a second time (test 5). Reductive dechlorination of TCFE primarily to cis-DCFE occurred
although trans-DCFE and CFE were also formed (Figure 4.Sd). TCFE was transformed at a
maximum rate of 0.047 day' between days 49 and 84 (t2). Co-injected formate was not
detected, indicating that formate-utilization and TCFE reductive dechlorination rates increased
together, as was observed in well I SC. Although relatively high concentrations of TCFE
persisted until the end of the test,
it
is of particular significance that TCFE reductive
dechlorination was stimulated where it had not initially been observed (test 1) (Figure 4.5c).
Well 21C is located upgradient from the contaminant plume in a location where
background contamination does not exist. Reductive dechlorination of TCFE did not occur
following the first injection of TCFE (test 1) (Figure 4.5e), which is consistent with the
expectation that an active population of dechlorinating organisms would not exist in the
absence of chlorinated compounds. After three successive additions of fumarate (tests 2-4),
TCFE was injected for a second time (test 5). Reductive dechlorination of TCFE primarily to
cis-DCFE occurred although trans-DCFE and CFE were also formed (Figure 4.50. TCFE
was transformed at a maximum rate of 0.14 day' between days 16 and 49 (Table 4.2). The
occurrence of TCFE reductive dechlorination in well 21C is of significance since it again
demonstrates that fumarate additions can stimulate reduction dechlorination even where it was
not initially observed.
Correlations between Fumarate and TCFE Transformation Behavior
Three types of fumarate transformation behavior were observed during successive
fumarate additions (tests 2-4). The first type of behavior is classified as that in which injected
fumarate was reduced to succinate at an increasing rate following each fumarate addition.
This type of behavior was observed in wells I OA and 9A. In well 1 OA, for example, fumarate
concentrations decreased more rapidly in test 4 than in test 2 (Figure 4.6a).
Likewise,
succinate formation was observed earlier in test 4 than in test 2. In both tests 2 and 4,
succinate concentrations eventually decreased to undetectable levels, indicating that microbial
populations utilized succinate. Based on results from laboratory experiments (9), it is likely
0.8
E
C
0
0
0
0
0.4
0
0.8
(b)21C
0.4
C
0
U
c_)
0
2
1
3
4
5
Time (Days)
0
2
4
6
8
10
Time (Days)
(c) 16C
0.8
Fumarate Test 2
Succinate Test 2
Fumarate Test 4
Succinate Test 4
0
0.4
0
0
S.
U
z
S.
--.4%
H00
0
2
4
6
8
10
Time (Days)
Figure 4.6. Tracer-normalized (TN) concentrations (see data analysis section) of
fumarate and its reduction product, succinate.
67
that succinate was utilized as an electron donor. The transformation behavior exhibited by
TCFE was similar to that exhibited by fumarate in that reductive dechlorination rates
increased between tests 1 and
5.
The similarities between TCFE and fumarate transformation
behaviors indicate that reductive dechlorination and fumarate reduction may be supported by
similar biogeochemical conditions in the A-zone.
The second type of fumarate transformation behavior is classified as that in which the
reduction of fumarate to succinate was not observed after its first addition (test 2) but was
observed after later additions (e.g. test 4). This type of behavior was observed in wells
and 21G.
1 5C
In well 21C, for example, fumarate concentrations decreased without the
concomitant detection of succinate in test 1 (Figure 4.6b). However, decreases in fumarate
concentrations were accompanied by succinate detection in test 4. While it is possible that
succinate was not detected in test 2 because it was formed as quickly as it was reduced, the
detection of succinate in test 4 indicates that microbial populations capable of reducing
fumarate were stimulated by successive fumarate additions.
The TCFE transformation
behavior observed in wells I SC and 21 C was consistent with that observed for fumarate since
reductive dechlorination was either observed only after a significant lag time or not observed
at all in these wells during test 1. Yet, reductive dechlorination was observed in these wells
during test
5.
This correlative behavior indicates that reductive dechlorination and fumarate
reduction may also be supported by similar biogeochemical conditions in the C-zone.
The third type of fumarate transformation behavior is classified as that in which the
reduction of fumarate to succinate was not observed during any of the successive fumarate
additions (tests 2-4). This type of behavior was observed in well 16C only (Figure 4.6c). This
behavior suggests that the biogeochemical conditions of well 1 6C were not conducive to the
stimulation of fumarate-reducing microbial populations. Based on the correlations between
reductive dechlorination and fumarate reduction observed in other wells, it is not surprising
that fumarate additions had a smaller effect on reductive dechlorination rates in well 16C than
in any other well.
Although TCFE and fumarate displayed parallel transformation behaviors in all five
wells at this site, the mechanism that caused TCFE reductive dechlorination rates to increase is
not known. A clue to understanding fumarate's role in enhancing reductive dechlorination
may be found in the fact that fumarate as well as its reduction product, succinate, were short-
lived in all wells except 16C (Figures 6a-c), where neither fumarate nor TCFE were
particularly susceptible to reduction. Because fumarate and succinate were short-lived, they
were not present in the groundwater during the final TCFE injection (test 5) and therefore
could not have been directly responsible for increased reductive dechlorination rates. Instead,
the consecutive additions of fumarate appeared to have altered the biogeochemical conditions
in a way that favored reductive dechlorination. Since the literature indicates that a number of
dechlorinating microorganisms utilize fumarate as an alternative electron acceptor (6-13) and
that at least one of them grows faster on fumarate than on chlorinated ethenes (11), it is
possible that fumarate additions stimulated the growth of dechiorinating/fumarate-reducing
microorganisms. This change in microbial community structure could have been responsible
for the enhancement of reduction rates for both TCFE and fumarate. It is also possible that
changes in microbial populations were induced by the presence of succinate (formed in situ),
which can be used as an electron donor (9), or by fumarate acting as an electron donor instead
of as an electron acceptor (14).
Another potential scenario is that by acting as an electron donor, fumarate reduced the
concentrations of indigenous electron acceptors such as oxygen or sulfate that can compete
with chlorinated ethenes. However, measured sulfate concentrations gave no indication that
sulfate reduction was occurring during these tests (data not shown). In some cases, TCFE
additions may also have contributed to increased TCFE reductive-dechlorination rates by
stimulating the growth of dechiorinating microorganisms. For example, during test 5 in well
bA, where background chlorinated ethene concentrations were negligible, changes in TCFE
reductive-dechlorination rates during the test followed a characteristic growth pattern with a
lag period of about 30 days. In contrast, during test 5 in well 9A, TCFE concentrations
decreased rapidly at the onset of the test, indicating that dechiorinating populations were
already present in the well at the beginning of the test. Accelerating rates, which can indicate
growth, were also observed during test
in well 15C and test 5 in well 21C. On-going
research in our group with soil microcosms is designed to further characterize the relation
I
between fumarate, TCFE, and TCE reductive dechlorination.
CONCLUSIONS
In this communication, a methodology for quantifying changes in in situ reductive
dechlorination rates due to a chemical amendment is described and demonstrated.
The
methodology is significant because progress in the development of bioremediation
technologies depends on an ability to measure the effectiveness of bioremediation techniques
69
in the field. To the best of our knowledge, this communication also describes the first use of
fumarate in a field method designed to enhance reductive dechlorination rates. TCFE
reductive dechlorination rates increased from 8.9 to 120 times in wells where reductive
dechlorination was initially observed. Reductive dechlorination was stimulated in wells where
it was not initially observed. Similarities in the transformation behaviors of TCFE and
fumarate indicated that reductive dechlorination and fumarate reduction were supported by
similar biogeochemical conditions.
ACKNOWLEDGEMENTS
We thank Timothy Buscheck, Kirk O'Reilly, Ralph Reed, and Mark Dolan for their
contributions. Kevin Tam, Robert Alumbaugh, Brian Davis, Jesse Jones, and Angelito Tirona
assisted in the field. This project was funded in part by grant number 1P42 ES 10338 from the
National Institute of Environmental Health Sciences (NIEHS), with funds from the U.S.
Environmental Protection Agency. The Western Region Hazardous Substance Research
Center and the ChevronTexaco Energy Research and Technology Company provided
additional support.
LITERATURE CITED
(1)
Bradley, P. M. Hydrogeol. J 2000, 8, 104-111.
(2)
McCarty, P. L. Science 1997, 276, 1521.
(3)
Vogel, T. M.; Criddle, C. S.; McCarty, P. L. Environ. Sci. Technol. 1987, 21, 722-
(4)
Lee, M. D.; Quinton, G. E.; Beeman, R. E.; Biehle, A. A.; Liddle, R. L.; Ellis, D. E.;
Jr., B. R. J. .1 md. Microbiol. Biotechnol. 1997, 18, 106-115.
(5)
Lee, M. D.; Odom, J. M.; Buchanan, R. J. J. Ann. Rev. Microbiol. 1998, 52, 423-452.
(6)
Neumann, A.; Scholz-Muramatsu, H.; Diekert, G. Arch. Microbiol. 1994, 162, 295-
(7)
Neumann, A.; Wohlfarth, G.; Diekert, G. Arch. Microbiol. 1995, 163, 276-281.
(8)
736.
301.
Scholz-Muramatsu, H.; Neumann, A.; Mel3mer, M.; Moore, E.; Diekert, G. Arch.
Microbiol. 1995, 163, 48-56.
(9)
Gerritse, J.; Renard, V.; Gomes, T. M. P.; Lawson, P. A.; Collins, M. D.; Gottschal, J.
C.Arch. Microbiol. 1996, 165, 132-140.
(10)
Miller, E.; Wohlfarth, G.; Diekert, G.Arch. Microbiol. 1997, 168, 513-5 19.
70
(ii)
Gerritse, J.; Drzyzga, 0.; Kloetstra, 0.; Keijmel, M.; Wiersum, L. P.; Flutson, R.;
(12)
Krumholz, L. R.; Sharp, R.; Fishbain, S. S. App!. Envir. Microbiol. 1996, 62, 4108-
(13)
Krumholz, L. R. mt. .1 Syst. Bacteriol. 1997, 47, 1262-1263.
(14)
Kengen, S. W. M.; Breidenbach, C. G.; Feiske, A.; Stams, A. J. M.; Schraa, 0.; Vos,
W. M. d. App!. Envir. Microbiol. 1999, 65, 2313-2316.
(15)
EPA, U. S. "Technical Protocol for Evaluating Natural Attentuation of Chlorinated
Solvents in Ground Water," 1998 EPA/600/R-98/128.
(16)
Washington, J. W.; Cameron, B. A. Environ. Toxico!. Chem. 2001, 20, 1909-19 15.
(17)
Chapelle, F. H.; Lovely, D. R. Appi. Environ. Microbiol. 1990, 56, 1865-1874.
(18)
Hageman, K. J.; Istok, J. D.; Field, J. A.; Buscheck, T. E.; Semprini, L. Environ. Sci.
Technol. 2001, 35, 1729-1735.
(19)
Istok, J. D.; Humphrey, M. D.; Schroth, M. H.; Hyman, M. R.; O'Reilly, K. T. Ground
Water 1997, 35, 619-631.
(20)
Vancheeswaran, S.; Hyman, M. R.; Semprini, L. Environ. Sci. Technol. 1999, 33,
(21)
Hageman, K. J.; Field, J. A.; Istok, J. D.; Schroth, M. H. submitted.
(22)
Buscheck, T. E.; O'Reilly, K. T.; Hickman, G. In In Situ and on-Site Bioremediation;
Battelle Press: Columbus, OH, 1997; Vol. 3, pp 149-154.
(23)
Buscheck, T. E. "Intrinsic Anaerobic Biotransformation of Chlorinated Solvents at a
Manufacturing Plant," Chevron Research and Technology Company, 1998.
(24)
Haggerty, R.; Schroth, M. H.; Istok, J. D. Ground Water 1998, 36, 314-324.
(25)
Snodgrass, M. F.; Kitanidis, P. K. Ground Water 1998, 36, 645-650.
(26)
Schnoor, J. L. Environmental Modeling: Fate and Transport of Pollutants in Water,
Air, and Soil; John Wiley & Sons, Inc.: New York, 1996.
(27)
(28)
Collins, M.D.; Gottschal, J. C.Appl. Envir. Microbiol. 1999, 65, 5212-5221.
4113.
2040-2045.
Syracuse Research Corporation "Estimations Programs Interface Suite," 2000,
www.epa.gov/oppt/exposure/docs!episuite.htm.
Pon, G.; Semprini, L. Published Abstract for In Situ and On-site Bioremediation
Seventh International Symposium: Orlando, FL, 2003.
71
CHAPTER 5: SUMMARY
Methods for measuring in situ transformation rates are needed to assess the potential
for natural attenuation and to evaluate engineered bioremediation approaches. However, in
situ transformation rates are difficult to measure because both transformation and transport
processes affect solute concentrations in groundwater.
This dissertation describes the
development and demonstration of a method for measuring in situ reductive dechlorination
rates in TCE-contaminated groundwater using single-well, "push-pull" tests.
One novel component of the rate-determination method presented in this dissertation
is the use of trichlorofluoroethene (TCFE) as a fluorinated surrogate for TCE. TCFE plays a
critical role in the method because it makes it possible to interrogate a TCE-contaminated
aquifer with an injected test solution. It would have been impractical to use TCE itself in the
injected test solution because injected and background TCE are not distinguishable. TCFE
was selected as a surrogate for TCE based on results from field experiments conducted in
TCE-contaminated groundwater at the site of a former chemical manufacturing plant in the
San Francisco Bay area.
Results from field experiments indicated that TCFE and TCE
experienced similar transport behavior in two distinct aquifer zones. Results from another set
of field experiments indicated that TCFE underwent reductive dechlorination by a pathway
analogous to that of TCE while retaining the fluorine label.
The second novel component of the rate-determination method presented in this
dissertation is the "forced mass balance" (FMB) data analysis technique. Although a data
analysis technique designed to determine in situ transformation rates for nonsorbing solutes
from push-pull test data had been described previously, the sorptive behavior of TCFE in the
selected aquifer made it impossible to interpret TCFE data with that technique. Hence, the
development of the FMB technique was critical because it made it possible to measure in situ
transformation rates of sorbing solutes, such as TCFE, from push-pull test data. The FMB
technique was evaluated by quantifying errors in rates derived by applying FMB to push-pull
test data generated by a numerical model. Results from the evaluation indicated that errors in
rates derived using FMB increase as the test duration, groundwater velocity, and ratio of
reactant to product retardation factor increase. In addition, errors in derived rates increase as
the reaction rate constant and aquifer dispersivity decrease. However, the error analysis also
72
indicated that a TCFE reductive dechlorination rate measured at the selected site was
underestimated relative to the true in situ rate by only 10%.
Finally, the utility of the rate-determination method was demonstrated by using it to
quantify changes in in situ reductive dechlorination rates due to the chemical amendment,
fumarate (trans-1,2-ethenedicarboxylate). TCFE transformation rates were measured before
and after three consecutive additions of fumarate in five wells. In wells where TCFE
reductive dechlorination was observed before fumarate additions, first-order rate constants
increased by factors ranging from 8.9 to
On the other hand, TCFE reductive
dechlorination was stimulated by fumarate additions in wells where it was not initially
120.
observed. To the author's best knowledge, this is the first time that fumarate was used in field
experiments to enhance reductive dechlorination rates.
73
BIBLIOGRAPHY
Abelson, P. H. Science 1990, 250, 733.
Abelson, P. H. Science 1990, 247, 141.
Arp, D. J.; Yeager, C. M.; Hyman, M. R. Biodegradation 2001, 12, 8 1-103.
Bekins, B.; Rittmann, B. E.; MacDonald, J. A. Eos 2001, 82, 57-58.
Bouwer, E. J. In Handbook of Bioremediation; RD Norris, Hinchee, RE, Brown, R, McCarty,
PL, Semprini, L, Wilson, DH, Kampbell, M, Reinhard, EG, Bouwer, R, Borden, C, Vogel,
TM, Thomas, JM, Ward, CH, Eds.; Lewis Publishers: Boca Raton, 1994; pp 149-175.
Bradley, P. M. Hydrogeol. J. 2000, 8, 104-111.
Bradley, P. M.; Chapelle, F. H. Environ. Sci. Technol. 2000, 34, 276 1-2763.
Bradley, P. M.; Chapelle, F. H. Environ. Sci. Technol. 1999, 33, 653-656.
Bradley, P. M.; Chapelle, F. H. Environ. Sci. Technol. 1999, 33, 3473-3476.
Buscheck, T. E. "Intrinsic Anaerobic Biotransformation of Chlorinated Solvents at a
Manufacturing Plant," Chevron Research and Technology Company, 1998.
Buscheck, T. E.; Alcantar, C. M. In Intrinsic Bioremediation; Robert E. Hinchee, Wilson,
John T., Downey, Douglas C., Eds.; Battelle Press: Columbus, 1995; pp 109-116.
Buscheck, T. E.; O'Reilly, K. T.; Hickman, G. In In Situ and on-Site Bioremediation; Battelle
Press: Columbus, OH, 1997; Vol. 3, pp 149-154.
California EPA Office of Environmental Health Hazard Assessment "Public Health Goals for
Trichloroethylene in Drinking Water," 1999, www.oehha.org/water/phgIpdf/tcef.pdf.
Chapelle, F. H. Ground-Water Microbiology and Geochemistry; John Wiley and Sons, Inc.:
New York, 1993.
Chapelle, F. H.; Lovely, D. R.Appl. Environ. Microbiol. 1990, 56, 1865-1874.
de Bruin, W.; Kotterman, M.; Posthumus, M.; Schraa, G.; Zehnder, A. App!. Environ.
Microbiol. 1992, 58, 1996-2000.
Domenico, P. A.; Schwartz, F. W. Physical and Chemical Hydrogeology; John Wiley and
Sons: New York, 1990.
Ensley, B. D. Ann. Rev. Microbiol. 1991, 45, 283-299.
74
Fantroussi, S. E.; Naveau, H.; Agathos, S. N. Biotechnol. Frog. 1998, 14, 167-188.
Fetter, C. W. Applied Hydrogeology; Prentice Hall: Upper Saddle River, New Jersey, 1994.
Gerritse, J.; Renard, V.; Gomes, T. M. P.; Lawson, P. A.; Collins, M. D.; Gottschal, J. C.
Arch. Microbiol. 1996, 165, 132-140.
Glod, G.; Angst, W.; Hollinger, C.; Schwarzenbach, R. P. Environ. Sci. Technol. 1997, 31,
253-260.
Hageman, K. J.; Field, J. A.; Istok, J. D.; Schroth, M. H. submitted.
Hageman, K. J.; Istok, J. D.; Field, J. A.; Buscheck, T. E.; Semprini, L. Environ.
2001, 35, 1729-1735.
Sd.
Technol.
1-laggerty, R.; Schroth, M. H.; Istok, J. D. Ground Water 1998, 36, 314-324.
J-lansch, C.; Leo, A. Substituent Constants for Correlation Analysis in Chemistry and Biology;
John Wiley and Sons, Inc.: New York, 1979.
Holliger, C.; Hahn, D.; Harmsen, H.; Ludwig, W.; Schumacher, W.; Tindall, B.; Vazquez, F.;
Weiss, N.; Zehnder, A. Arch. Microbiol. 1998, 169, 3 13-321.
Hollinger, C.; Wohlfarth, G.; Diekert, G. FEMSMicrobiol. Rev. 1998, 22, 383-398.
Istok, J. D.; Humphrey, M. D.; Schroth, M. H.; Hyman, M. R.; O'Reilly, K. T. Ground Water
1997, 35, 619-63 1.
Kaster, M. Appl. Environ. Microbiol. 1991, 57, 2039-2045.
Kengen, S. W. M.; Breidenbach, C. G.; Feiske, A.; Stams, A. J. M.; Schraa, G.; Vos, W. M. d.
Appl. Envir. Microbiol. 1999, 65, 2313-2316.
Kleikemper, J.; Schroth, M. H.; Sigler, W. V.; Schmucki, M.; Bernasconi, S. M.; Zeyer, J.
Appl. Environ. Microbiol. 2002, 68, 1516-1523.
Krumholz, L. R. mt. i Syst. Bacteriol. 1997, 47, 1262-1263.
Krumholz, L. R.; Sharp, R.; Fishbain, S. S. Appi. Envir. Microbiol. 1996, 62, 4108-4113.
Landa, A.; Sipkema, E.; Weijma, J.; Beenackers, A.; Dolfing, J.; Janssen, D. Appl. Environ.
Microbiol. 1994, 60, 3368-3374.
Lee, M. D.; Odom, J. M.; Buchanan, R. J. J. Ann. Rev. Microbiol. 1998, 52, 423-452.
Lee, M. D.; Quinton, G. E.; Beeman, R. E.; Biehle, A. A.; Liddle, R. L.; Ellis, D. E.; Jr., B. R.
J. I md. Microbiol. Biotechnol. 1997, 18, 106-115.
Madsen, E. L. Environ. Sci. Technol. 1991, 25, 1662-1673.
75
Major, D. W.; Hodgins, E. W.; Butler, B. J. In On-Site Bioreclamation Processes for
Xenobiotic and Hydrocarbon Treatment; Robert E. Hinchee, Olfenbuttel, Robert F., Eds.;
Butterworth-Heinemann: Boston, 1991.
Mars, A. E.; Houwing, J.; Dolfing, J.; Janssen, D. B. Appi. Environ. Microbiol. 1996, 62, 886891.
Maymó-Gatell, X.; Tandoi, V.; Gossett, J.; Zinder, S. App!. Environ. Microbiol. 1995, 64,
3928-3933.
Maymó-GatelI, X.; Chien, Y.; Gossett, J. M.; Zinder, S. H. Science 1997, 276, 1568-1571.
McAllister, P.M.; Chiang, C. Y. Ground Water Monit. Rev. 1994, 14, 161-173.
McCarty, P. L. Science 1997, 276, 152 1-1522.
McCarty, P. L.; Semprini, L. In Handbook of Bioremediation; RD Norris, Hinchee, RE,
Brown, R, McCarty, PL, Semprini, L, Wilson, DH, Kampbell, M, Reinhard, EG, Bouwer, R,
Borden, C, Vogel, TM, Thomas, JM, Ward, CH, Eds.; Lewis Publishers: Boca Raton, 1994;
pp 87-116.
McCarty, P. L.; Wilson, J. T. In Biorem. Haz. Wastes; Environmental Protection Agency
Office of Research and Development, Ed.: Washington, D.C., 1992; pp 68-72.
McNab, W. W. J.; Dooher, B. P. Ground Water 1998, 36, 983-987.
Miller, E.; Wohlfarth, G.; Diekert, G. Arch. Microbio!. 1997, 168, 513-519.
Mohn, W. W.; Tiedje, J. M. Microbiol. Rev. 1992, 56, 482-507.
National Research Council In Situ Bioremediation: When Does It Work?; National Academy
Press: Washington, D.C., 1993.
Neumann, A.; Scholz-Muramatsu, H.; Diekert, G. Arch. MicrobioL 1994, 162, 295-301.
Neumann, A.; Wohlfarth, G.; Diekert, G. Arch. Microbio!. 1995, 163, 276-281.
Pombo, S. A.; Pelz, 0.; Schroth, M. H.; Zeyer, J. FEMSMicrobio!. Eco!. 2002, 41, 259-267.
Pon, G.; Semprini, L. Published Abstract for In Situ and On-site Bioremediation Seventh
International Symposium: Orlando, FL, 2003.
Schnoor, J. L. Environmental Modeling: Fate and Transport of Pollutants in Water, Air, and
Soil; John Wiley & Sons, Inc.: New York, 1996.
Scholz-Muramatsu, H.; Neumann, A.; Mel3mer, M.; Moore, E.; Diekert, G. Arch. Microbiol.
1995, 163, 48-56.
76
Schroth, M. H.; Istok, J. D.; Conner, G. T.; Hyman, M. R.; Haggerty, R.; O'Reilly, K. T.
Ground Water 1998, 36, 924-937.
Schroth, M. H.; Kleikemper, J.; Bollinger, C.; Bernasconi, S. M.; Zeyer, J. I Contam. Hydrol.
2001, 51, 179-195.
Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental Organic Chemistry;
John Wiley and Sons, Inc.: New York, 1993.
Semprini, L.; Kitandis, P. K.; Kampbell, D. H.; Wilson, J. T. Water Resour. Res. 1995, 31,
105 1-1062.
Snodgrass, M. F.; Kitanidis, P. K. Ground Water 1998, 36, 645-65 0.
Suarez, M. P.; Rifai, H. S. Biorem. 1 1999, 3, 337-362.
Syracuse Research Corporation "Estimations
www.epa.gov/oppt/exposure/docs/episuite.htm.
Programs
Interface
Suite,"
2000,
Travis, C. C.; Doty, C. B. Environ. Sci. Technol. 1990, 24, 1464-1466.
US EPA Office of Groundwater
and Drinking Water "Consumer Factsheet on
Trichloroethylene," 2003, www.epa.gov/OGWDW.
US EPA Office of Research and Development "Technical Protocol for Evaluating Natural
Attenuation of Chlorinated Solvents in Ground Water," 1998 EPA/600/R-98/128.
Vancheeswaran, S.; Hyman, M. R.; Semprini, L. Environ. Sci. Technol. 1999, 33, 2040-2045.
Vogel, T. M. In Handbook of Bioremediation; RD Norris, Hinchee, RE, Brown, R, McCarty,
PL, Semprini, L, Wilson, DH, Kampbell, M, Reinhard, EG, Bouwer, R, Borden, C, Vogel,
TM, Thomas, JM, Ward, CH, Eds.; Lewis Publishers: Boca Raton, 1994; pp 20 1-225.
Vogel, T. M.; Criddle, C. S.; McCarty, P. L. Environ. Sci. Technol. 1987, 21, 722-736.
Washington, J. W.; Cameron, B. A. Environ. Toxicol. Chem. 2001, 20, 1909-1915.
Weaver, J. W.; Wilson, J. 1.; Kampbell, D. H. Proceedings of the Symposium on Natural
Attenuation of Chlorinated Organics in Ground Water; Office of Research and Development,
Ed.; U.S. EPA: Dallas, Texas, 1997; pp 7 1-75.
White, M. D.; Oostrom, M. "Stomp: Subsurface Transport over Multiple Phases. User's
Guide," Pacific Northwest National Laboratory, 1997 PNNL-12034.
Wiedemeier, T. H.; Swanson, M. A.; Moutoux, D. E.; Wilson, J. T.; Kampbell, D. H.; Hansen,
J. E.; Haas, P. In Symposium on Natural Attenuation of Chlorinated Organics in Ground
Water: Dallas, TX, 1996; pp 35-59.
'14
Wiedemeier, T. H.; Swanson, M. A.; Wilson, J. T.; Kampbell, D. H.; Miller, R. N.; Hansen, J.
E. Ground Water Monit. Rev. 1996, 16, 186-194.
Download